• Neural Implicit Heart Coordinates (NIHCs) proposed as a standardized anatomical reference system. • Dual-network model predicts NIHCs from sparse segmentations without requiring 3D meshes. • Method accurately reconstructs biventricular heart anatomy, including the four valve annuli. • Extensive evaluation on over 10,000 cases spanning both healthy and diseased populations. Accurate reconstruction of cardiac anatomy from sparse clinical images remains a major challenge in patient-specific modeling. While neural implicit functions have previously been applied to this task, their application to mapping anatomical consistency across subjects has been limited. In this work, we introduce Neural Implicit Heart Coordinates (NIHCs), a standardized implicit coordinate system, based on universal ventricular coordinates, that provides a common anatomical reference frame for the human heart. Our method predicts NIHCs directly from a limited number of 2D segmentations (sparse acquisition) and subsequently decodes them into dense 3D segmentations and high-resolution meshes at arbitrary output resolution. Trained on a large dataset of 5,000 cardiac meshes, the model achieves high reconstruction accuracy on clinical contours, with mean Euclidean surface errors of 2.51 ± 0.33 mm in a diseased cohort (n=4549) and 2.31 ± 0.36 mm in a healthy cohort (n=5576). The NIHC representation enables anatomically coherent reconstruction even under severe slice sparsity and segmentation noise, faithfully recovering complex structures such as the valve planes. Compared with traditional pipelines, inference time is reduced from over 60 s to 5–15 s. These results demonstrate that NIHCs constitute a robust and efficient anatomical representation for patient-specific 3D cardiac reconstruction from minimal input data.

Image registration is fundamental in medical imaging applications, such as disease progression analysis or radiation therapy planning. The primary objective of image registration is to precisely capture the deformation between two or more images, typically achieved by minimizing an optimization problem. Due to its inherent ill-posedness, regularization is a key component in driving the solution toward anatomically meaningful deformations. A wide range of regularization methods has been proposed for both conventional and deep learning-based registration. However, the appropriate application of regularization techniques often depends on the specific registration problem, and no "one-fits-all" method exists. Despite its importance, regularization is often overlooked or addressed with default approaches, assuming existing methods are sufficient. A comprehensive and structured review remains missing. This review addresses this gap by introducing a novel taxonomy that systematically categorizes the diverse range of proposed regularization methods. It highlights the emerging field of learned regularization, which leverages data-driven techniques to automatically derive deformation properties from the data. Moreover, this review examines the transfer of regularization methods from conventional to learning-based registration, identifies open challenges, and outlines future research directions. By emphasizing the critical role of regularization in image registration, we hope to inspire the research community to reconsider regularization strategies in modern registration algorithms and to explore this rapidly evolving field further.

Reliable recognition and localization of surgical instruments in endoscopic video recordings are foundational for a wide range of applications in computer- and robot-assisted minimally invasive surgery (RAMIS), including surgical training, skill assessment, and autonomous assistance. However, robust performance under real-world conditions remains a significant challenge. Incorporating surgical context - such as the current procedural phase - has emerged as a promising strategy to improve robustness and interpretability. To address these challenges, we organized the Surgical Procedure Phase, Keypoint, and Instrument Recognition (PhaKIR) sub-challenge as part of the Endoscopic Vision (EndoVis) challenge at MICCAI 2024. We introduced a novel, multi-center dataset comprising thirteen full-length laparoscopic cholecystectomy videos collected from three distinct medical institutions, with unified annotations for three interrelated tasks: surgical phase recognition, instrument keypoint estimation, and instrument instance segmentation. Unlike existing datasets, ours enables joint investigation of instrument localization and procedural context within the same data while supporting the integration of temporal information across entire procedures. We report results and findings in accordance with the BIAS guidelines for biomedical image analysis challenges. The PhaKIR sub-challenge advances the field by providing a unique benchmark for developing temporally aware, context-driven methods in RAMIS and offers a high-quality resource to support future research in surgical scene understanding.

Neural implicit k-space representations (NIK) have shown promising results for dynamic magnetic resonance imaging (MRI) at high temporal resolutions. Yet, reducing acquisition time, and thereby available training data, results in severe performance drops due to overfitting. To address this, we introduce a novel self-supervised k-space loss function L PISCO , applicable for regularization of NIK-based reconstructions. The proposed loss function is based on the concept of parallel imaging-inspired self-consistency (PISCO), enforcing a consistent global k-space neighborhood relationship without requiring additional data. Quantitative and qualitative evaluations on static and dynamic MR reconstructions show that integrating PISCO significantly improves NIK representations, making it a competitive dynamic reconstruction method without constraining the temporal resolution. Particularly at high acceleration factors (R ≥ 50), NIK with PISCO can avoid temporal oversmoothing of state-of-the-art methods and achieves superior spatio-temporal reconstruction quality. Furthermore, an extensive analysis of the loss assumptions and stability shows PISCO’s potential as versatile self-supervised k-space loss function for further applications and architectures. Code is available at: https://github.com/compai-lab/2025-pisco-spieker

Cardiac Diffusion Tensor Imaging (cDTI) provides unique insights into myocardial microstructure in-vivo but requires averaging multiple repetitions for adequate signal quality, leading to prohibitively long acquisition times. Standard acceleration strategies, such as reducing repetitions and employing simultaneous multi-slice (SMS) imaging, are limited by low signal-to-noise ratio (SNR) and inter-slice leakage artefacts, respectively. We introduce ORCAS, a unified framework that synergistically combines a novel variable CAIPIRINHA acquisition with an artefact-aware AI reconstruction to overcome these challenges. The variable CAIPIRINHA scheme decoheres SMS artefacts across repetitions, while our dual-domain deep learning model simultaneously suppresses these artefacts and combats the low SNR from fewer repetitions. The model is guided by patient-specific single-band auxiliary data to preserve anatomical fidelity. Validated on ex-vivo hearts with and without anomalies, ORCAS achieves an over 18-fold acceleration by combining these strategies, reducing a whole-heart scan from over two hours to under 7 min. This is accomplished while reducing errors in key biomarkers, such as Fractional Anisotropy, by up to 64%. The framework preserves essential microstructural properties and the delineation of abnormalities, representing a significant step towards the clinical translation of whole-heart cDTI. • Novel variable CAIPIRINHA reduces SMS artefacts in cardiac DTI. • AI framework achieves 18×acceleration while preserving biomarkers. • Reduces DTI errors by 64% compared to conventional reconstruction. • Enables whole-heart cDTI in under 7 min vs over 2 h. • Preserves abnormalities even at extreme acceleration factors.

Brain lesion segmentation is critical for diagnosing and monitoring neurological diseases such as Multiple Sclerosis (MS). However, lesion variability and differences in scanners and acquisition techniques pose a significant challenge to the robust generalization of automated segmentation models beyond their training domain. Traditional augmentations, such as rotation, intensity shifts, and scalings, often fail to capture the wide diversity observed across patient cases, limiting model generalizability. Random Convolutions (RC) address this limitation by introducing diverse intensity variations while preserving anatomical structures. Using an nnUNet-based model enhanced with RC augmentations, we achieved 5th place in the MSLesSeg challenge, highlighting that RC augmentations offer competitive in-domain performance. Building on this, we further assess model performance, both in terms of lesion detection and segmentation, in- and out-of-domain. We compare RC with several state-of-the-art augmentation and domain generalization strategies and show that an nnUNet trained with the RC augmentation is competitive in-domain and demonstrates superior generalization performance. • RC augmentation enhances out-of-domain performance in MS lesion detection and segmentation. • The model achieves a top-5 ranking in the MSLesSeg challenge with RC-enhanced nnUNet . • RC improves detection of small lesions, particularly under domain shift. • RC achieves optimal performance with mid-sized kernels ( k = 5–7) and moderate layer depths ( L = 6–8). • RC is a simple yet effective strategy for robust MS lesion segmentation.

Abstract Fine-tuning large language models (LLMs) with limited data poses a practical challenge in low-resource languages, specialized domains, and constrained deployment settings. While pre-trained LLMs provide strong foundations, effective adaptation under data scarcity requires focused and efficient fine-tuning techniques. This paper presents a structured and practical survey of recent methods for fine-tuning LLMs in data-scarce scenarios. We systematically review parameter-efficient fine-tuning techniques that lower training and deployment costs, domain and cross-lingual adaptation methods for both encoder and decoder models, and model specialization strategies. We further examine preference alignment approaches that guide model behavior using limited human or synthetic feedback, emphasizing sample and compute efficiency. Throughout, we highlight empirical trade-offs, selection criteria, and best practices for choosing suitable techniques based on task constraints, including model scaling, data scaling, and the mitigation of catastrophic forgetting. The aim is to equip researchers and practitioners with actionable insights for effectively fine-tuning LLMs when data and resources are limited.

Coronary artery stenosis is a leading cause of cardiovascular disease, diagnosed by analyzing the coronary arteries from multiple angiography views. Although numerous deep-learning models have been proposed for stenosis detection from a single angiography view, their performance heavily relies on expensive view-level annotations, which are often not readily available in hospital systems. Moreover, these models fail to capture the temporal dynamics and dependencies among multiple views, which are crucial for clinical diagnosis. To address this, we propose SegmentMIL, a transformer-based multi-view multiple-instance learning framework for patient-level stenosis classification. Trained on a real-world clinical dataset, using patient-level supervision and without any view-level annotations, SegmentMIL jointly predicts the presence of stenosis and localizes the affected anatomical region, distinguishing between the right and left coronary arteries and their respective segments. SegmentMIL obtains high performance on internal and external evaluations and outperforms both view-level models and classical MIL baselines, underscoring its potential as a clinically viable and scalable solution for coronary stenosis diagnosis. Our code is available at https://github.com/NikolaCenic/mil-stenosis.

Magnetic resonance imaging of fetal and neonatal brains reveals rapid neurodevelopment marked by substantial anatomical changes unfolding within days. Studying this critical stage of the developing human brain, therefore, requires accurate brain models-referred to as atlases-of high spatial and temporal resolution. To meet these demands, established traditional atlases and recently proposed deep learning-based methods rely on large and comprehensive datasets. This poses a major challenge for studying brains in the presence of pathologies for which data remains scarce. We address this limitation with CINeMA (Conditional Implicit Neural Multi-Modal Atlas), a novel framework for creating high-resolution, spatio-temporal, multimodal brain atlases, suitable for low-data settings. Unlike established methods, CINeMA operates in latent space, avoiding compute-intensive image registration and reducing atlas construction times from days to minutes. Furthermore, it enables flexible conditioning on anatomical features including gestational age, birth age, and pathologies like agenesis of the corpus callosum and ventriculomegaly of varying degree. CINeMA supports downstream tasks such as tissue segmentation and age prediction whereas its generative properties enable synthetic data creation and anatomically informed data augmentation. Surpassing state-of-the-art methods in accuracy, efficiency, and versatility, CINeMA represents a powerful tool for advancing brain research. We release the code and atlases at https://github.com/m-dannecker/CINeMA.

Longitudinal brain analysis is essential for understanding healthy aging and identifying pathological deviations. Longitudinal registration of sequential brain MRI underpins such analyses. However, existing methods are limited by reliance on densely sampled time series, a trade-off between accuracy and temporal smoothness, and an inability to prospectively forecast future brain states. To overcome these challenges, we introduce TimeFlow, a learning-based framework for longitudinal brain MRI registration. TimeFlow uses a U-Net backbone with temporal conditioning to model neuroanatomy as a continuous function of age. Given only two scans from an individual, TimeFlow estimates accurate and temporally coherent deformation fields, enabling non-linear extrapolation to predict future brain states. This is achieved by our proposed inter-/extrapolation consistency constraints applied to both the deformation fields and deformed images. Remarkably, these constraints preserve temporal consistency and continuity without requiring explicit smoothness regularizers or densely sampled sequential data. Extensive experiments demonstrate that TimeFlow outperforms state-of-the-art methods in terms of both future timepoint forecasting and registration accuracy. Moreover, TimeFlow supports novel biological brain aging analyses by differentiating neurodegenerative trajectories from normal aging without requiring segmentation, thereby eliminating the need for labor-intensive annotations and mitigating segmentation inconsistency. TimeFlow offers an accurate, data-efficient, and annotation-free framework for longitudinal analysis of brain aging and chronic diseases, capable of forecasting brain changes beyond the observed study period.

Anatomical atlases are widely used for population studies and analysis. Conditional atlases target a specific sub-population defined via certain conditions, such as demographics or pathologies, and allow for the investigation of fine-grained anatomical differences like morphological changes associated with ageing or disease. Existing approaches use either registration-based methods that are often unable to handle large anatomical variations or generative adversarial models, which are challenging to train since they can suffer from training instabilities. Instead of generating atlases directly in as intensities, we propose using latent diffusion models to generate deformation fields, which transform a general population atlas into one representing a specific sub-population. Our approach ensures structural integrity, enhances interpretability and avoids hallucinations that may arise during direct image synthesis by generating this deformation field and regularising it using a neighbourhood of images. We compare our method to several state-of-the-art atlas generation methods using brain MR images from the UK Biobank. Our method generates highly realistic atlases with smooth transformations and high anatomical fidelity, outperforming existing baselines. We demonstrate the quality of these atlases through comprehensive evaluations, including quantitative metrics for anatomical accuracy, perceptual similarity, and qualitative analyses displaying the consistency and realism of the generated atlases.

Supervised contrastive learning (SupCon) has proven to be a powerful alternative to the standard cross-entropy loss for classification of multi-class balanced datasets. However, it struggles to learn well-conditioned representations of datasets with long-tailed class distributions. This problem is potentially exacerbated for binary imbalanced distributions, which are commonly encountered during many real-world problems such as medical diagnosis. In experiments on seven binary datasets of natural and medical images, we show that the performance of SupCon decreases with increasing class imbalance. To substantiate these findings, we introduce two novel metrics that evaluate the quality of the learned representation space. By measuring the class distribution in local neighborhoods, we are able to uncover structural deficiencies of the representation space that classical metrics cannot detect. Informed by these insights, we propose two new supervised contrastive learning strategies tailored to binary imbalanced datasets that improve the structure of the representation space and increase downstream classification accuracy over standard SupCon by up to 35%. We make our code available.1

Current AI frameworks for brain decoding and encoding, typically train and test models within the same datasets. This limits their utility for cognitive training (neurofeedback) for which it would be useful to pool experiences across individuals to better simulate stimuli not sampled during training. A key obstacle to model generalisation is the degree of variability of inter-subject cortical organisation, which makes it difficult to align or compare cortical signals across participants. In this paper we address this through use of surface vision transformers, which build a generalisable model of cortical functional dynamics, through encoding the topography of cortical networks and their interactions as a moving image across a surface. This is then combined with tri-modal self-supervised contrastive (CLIP) alignment of audio, video, and fMRI modalities to enable the retrieval of visual and auditory stimuli from patterns of cortical activity (and vice-versa). We validate our approach on 7T task-fMRI data from 174 healthy participants engaged in the movie-watching experiment from the Human Connectome Project (HCP). Results show that it is possible to detect which movie clips an individual is watching purely from their brain activity, even for individuals and movies *not seen during training*. Further analysis of attention maps reveals that our model captures individual patterns of brain activity that reflect semantic and visual systems. This opens the door to future personalised simulations of brain function. Code \& pre-trained models will be made available at https://github.com/metrics-lab/sim.

Accurately estimating the informativeness of individual samples in a dataset is an important objective in deep learning, as it can guide sample selection, which can improve model efficiency and accuracy by removing redundant or potentially harmful samples. We propose $\text{\textit{Laplace Sample Information}}$ ($\mathsf{LSI}$) measure of sample informativeness grounded in information theory widely applicable across model architectures and learning settings. $\mathsf{LSI}$ leverages a Bayesian approximation to the weight posterior and the KL divergence to measure the change in the parameter distribution induced by a sample of interest from the dataset. We experimentally show that $\mathsf{LSI}$ is effective in ordering the data with respect to typicality, detecting mislabeled samples, measuring class-wise informativeness, and assessing dataset difficulty. We demonstrate these capabilities of $\mathsf{LSI}$ on image and text data in supervised and unsupervised settings. Moreover, we show that $\mathsf{LSI}$ can be computed efficiently through probes and transfers well to the training of large models.

Topological correctness plays a critical role in many image segmentation tasks, yet most networks are trained using pixel-wise loss functions, such as Dice, neglecting topological accuracy. Existing topology-aware methods often lack robust topological guarantees, are limited to specific use cases, or impose high computational costs. In this work, we propose a novel, graph-based framework for topologically accurate image segmentation that is both computationally efficient and generally applicable. Our method constructs a component graph that fully encodes the topological information of both the prediction and ground truth, allowing us to efficiently identify topologically critical regions and aggregate a loss based on local neighborhood information. Furthermore, we introduce a strict topological metric capturing the homotopy equivalence between the union and intersection of prediction-label pairs. We formally prove the topological guarantees of our approach and empirically validate its effectiveness on binary and multi-class datasets, demonstrating state-of-the-art performance with up to fivefold faster loss computation compared to persistent homology methods.

Robust validation metrics remain essential in contemporary deep learning, not only to detect overfitting and poor generalization, but also to monitor training dynamics. In the supervised classification setting, we investigate whether interactions between training data and model weights can yield such a metric that both tracks generalization during training and attributes performance to individual training samples. We introduce Gradient-Weight Alignment (GWA), quantifying the coherence between per-sample gradients and model weights. We show that effective learning corresponds to coherent alignment, while misalignment indicates deteriorating generalization. GWA is efficiently computable during training and reflects both sample-specific contributions and dataset-wide learning dynamics. Extensive experiments show that GWA accurately predicts optimal early stopping, enables principled model comparisons, and identifies influential training samples, providing a validation-set-free approach for model analysis directly from the training data.

Direct image-to-graph transformation is a challenging task that involves solving object detection and relationship prediction in a single model. Due to this task's complexity, large training datasets are rare in many domains, making the training of deep-learning methods challenging. This data sparsity necessitates transfer learning strategies akin to the state-of-the-art in general computer vision. In this work, we introduce a set of methods enabling cross-domain and cross-dimension learning for image-to-graph trans-formers. We propose (1) a regularized edge sampling loss to effectively learn object relations in multiple domains with different numbers of edges, (2) a domain adaptation frame-work for image-to-graph transformers aligning image- and graph-level features from different domains, and (3) a projection function that allows using 2D data for training 3D transformers. We demonstrate our method's utility in cross-domain and cross-dimension experiments, where we utilize labeled data from 2D road networks for simultaneous learning in vastly different target domains. Our method consistently outperforms standard transfer learning and self-supervised pretraining on challenging benchmarks, such as retinal or whole-brain vessel graph extraction.11Code: github.com/AlexanderHBerger/cross-dim_i2g

The Developing Human Connectome Project (dHCP) aims to explore developmental patterns of the human brain during the perinatal period. An automated processing pipeline has been developed to extract high-quality cortical surfaces from structural brain magnetic resonance (MR) images for the dHCP neonatal dataset. However, the current implementation of the pipeline requires more than 6.5 h to process a single MRI scan, making it expensive for large-scale neuroimaging studies. In this paper, we propose a fast deep learning (DL) based pipeline for dHCP neonatal cortical surface reconstruction, incorporating DL-based brain extraction, cortical surface reconstruction and spherical projection, as well as GPU-accelerated cortical surface inflation and cortical feature estimation. We introduce a multiscale deformation network to learn diffeomorphic cortical surface reconstruction end-to-end from T2-weighted brain MRI. A fast unsupervised spherical mapping approach is integrated to minimize metric distortions between cortical surfaces and projected spheres. The entire workflow of our DL-based dHCP pipeline completes within only 24 s on a modern GPU, which is nearly 1000 times faster than the original dHCP pipeline. The qualitative assessment demonstrates that for 82.5% of the test samples, the cortical surfaces reconstructed by our DL-based pipeline achieve superior (54.2%) or equal (28.3%) surface quality compared to the original dHCP pipeline.

Cardiovascular diseases (CVD) can be diagnosed using various diagnostic modalities. The electrocardiogram (ECG) is a cost-effective and widely available diagnostic aid that provides functional information of the heart. However, its ability to classify and spatially localise CVD is limited. In contrast, cardiac magnetic resonance (CMR) imaging provides detailed structural information of the heart and thus enables evidence-based diagnosis of CVD, but long scan times and high costs limit its use in clinical routine. In this work, we present a deep learning strategy for cost-effective and comprehensive cardiac screening solely from ECG. Our approach combines multimodal contrastive learning with masked data modelling to transfer domain-specific information from CMR imaging to ECG representations. In extensive experiments using data from 40,044 UK Biobank subjects, we demonstrate the utility and generalisability of our method for subject-specific risk prediction of CVD and the prediction of cardiac phenotypes using only ECG data. Specifically, our novel multimodal pre-training paradigm improves performance by up to 12.19% for risk prediction and 27.59% for phenotype prediction. In a qualitative analysis, we demonstrate that our learned ECG representations incorporate information from CMR image regions of interest. Our entire pipeline is publicly available at https://github.com/oetu/MMCL-ECG-CMR.

Cardiac magnetic resonance (CMR) imaging is the gold standard for non-invasive cardiac assessment, offering rich spatio-temporal views of the heart’s anatomy and physiology. Patient-level health factors, such as demographics, metabolic, and lifestyle, are known to substantially influence cardiovascular health and disease risk, yet remain uncaptured by CMR alone. To holistically understand cardiac health and to enable the best possible interpretation of an individual’s disease risk, CMR and patient-level factors must be jointly exploited within an integrated framework. Recent multi-modal approaches have begun to bridge this gap, yet they often rely on limited spatio-temporal data and focus on isolated clinical tasks, thereby hindering the development of a comprehensive representation for cardiac/health evaluation. To overcome these limitations, we introduce ViTa , a step toward foundation models that delivers a comprehensive representation of the heart and a precise interpretation of individual disease risk. Leveraging data from 42,000 UK Biobank participants, ViTa integrates 3D+T cine stacks from short-axis and long-axis views, enabling a complete capture of the cardiac cycle. These imaging data are then fused with detailed tabular patient-level factors, enabling context-aware insights. This multi-modal paradigm supports a wide spectrum of downstream tasks, including cardiac phenotype and physiological feature prediction, segmentation, and classification of cardiac/metabolic diseases within a single unified framework. By learning a shared latent representation that bridges rich imaging features and patient context, ViTa moves beyond traditional, task-specific models toward a universal, patient-specific understanding of cardiac health, highlighting its potential to advance clinical utility and scalability in cardiac analysis. 2

Deep learning-based medical image segmentation techniques have shown promising results when evaluated based on conventional metrics such as the Dice score or Intersection-over-Union. However, these fully automatic methods often fail to meet clinically acceptable accuracy, especially when topological constraints should be observed, e.g., continuous boundaries or closed surfaces. In medical image segmentation, the correctness of a segmentation in terms of the required topological genus sometimes is even more important than the pixel-wise accuracy. Existing topology-aware approaches commonly estimate and constrain the topological structure via the concept of persistent homology (PH). However, these methods are difficult to implement for high dimensional data due to their polynomial computational complexity. To overcome this problem, we propose a novel and fast approach for topology-aware segmentation based on the Euler Characteristic ( $\chi $ ). First, we propose a fast formulation for $\chi $ computation in both 2D and 3D. The scalar $\chi $ error between the prediction and ground-truth serves as the topological evaluation metric. Then we estimate the spatial topology correctness of any segmentation network via a so-called topological violation map, i.e., a detailed map that highlights regions with $\chi $ errors. Finally, the segmentation results from the arbitrary network are refined based on the topological violation maps by a topology-aware correction network. Our experiments are conducted on both 2D and 3D datasets and show that our method can significantly improve topological correctness while preserving pixel-wise segmentation accuracy.

Weakly supervised object detection (WSup-OD) increases the usefulness and interpretability of image classification algorithms without requiring additional supervision. The successes of multiple instance learning in this task for natural images, however, do not translate well to medical images due to the very different characteristics of their objects (i.e. pathologies). In this work, we propose Weakly Supervised ROI Proposal Networks (WSRPN), a new method for generating bounding box proposals on the fly using a specialized region of interest-attention (ROI-attention) module. WSRPN integrates well with classic backbone-head classification algorithms and is end-to-end trainable with only image-label supervision. We experimentally demonstrate that our new method outperforms existing methods in the challenging task of disease localization in chest X-ray images. Code: https://github.com/philip-mueller/wsrpn.

Biophysical modeling, particularly involving partial differential equations (PDEs), offers significant potential for tailoring disease treatment protocols to individual patients. However, the inverse problem-solving aspect of these models presents a substantial challenge, either due to the high computational requirements of model-based approaches or the limited robustness of deep learning (DL) methods. We propose a novel framework that leverages the unique strengths of both approaches in a synergistic manner. Our method incorporates a DL ensemble for initial parameter estimation, facilitating efficient downstream evolutionary sampling initialized with this DL-based prior. We showcase the effectiveness of integrating a rapid deep-learning algorithm with a high-precision evolution strategy in estimating brain tumor cell concentrations from magnetic resonance images. The DL-Prior plays a pivotal role, significantly constraining the effective sampling-parameter space. This reduction results in a fivefold convergence acceleration and a Dice-score of 95%.

We propose a self-supervised feature learning assisted reconstruction (SSFL-Recon) framework for MRI reconstruction to address the limitation of existing supervised learning methods. Although recent deep learning-based methods have shown promising performance in MRI reconstruction, most require fully-sampled images for supervised learning, which is challenging in practice considering long acquisition times under respiratory or organ motion. Moreover, nearly all fully-sampled datasets are obtained from conventional reconstruction of mildly accelerated datasets, thus potentially biasing the achievable performance. The numerous undersampled datasets with different accelerations in clinical practice, hence, remain underutilized. To address these issues, we first train a self-supervised feature extractor on undersampled images to learn sampling-insensitive features. The pre-learned features are subsequently embedded in the self-supervised reconstruction network to assist in removing artifacts. Experiments were conducted retrospectively on an in-house 2D cardiac Cine dataset, including 91 cardiovascular patients and 38 healthy subjects. The results demonstrate that the proposed SSFL-Recon framework outperforms existing self-supervised MRI reconstruction methods and even exhibits comparable or better performance to supervised learning up to ${16}\times $ retrospective undersampling. The feature learning strategy can effectively extract global representations, which have proven beneficial in removing artifacts and increasing generalization ability during reconstruction.

In differentially private (DP) machine learning, the privacy guarantees of DP mechanisms are often reported and compared on the basis of a single $(\varepsilon, \delta)$-pair. This practice overlooks that DP guarantees can vary substantially even between mechanisms sharing a given $(\varepsilon, \delta)$, and potentially introduces privacy vulnerabilities which can remain undetected. This motivates the need for robust, rigorous methods for comparing DP guarantees in such cases. Here, we introduce the $\Delta$-divergence between mechanisms which quantifies the worst-case excess privacy vulnerability of choosing one mechanism over another in terms of $(\varepsilon, \delta)$, $f$-DP and in terms of a newly presented Bayesian interpretation. Moreover, as a generalisation of the Blackwell theorem, it is endowed with strong decision-theoretic foundations. Through application examples, we show that our techniques can facilitate informed decision-making and reveal gaps in the current understanding of privacy risks, as current practices in DP-SGD often result in choosing mechanisms with high excess privacy vulnerabilities.

The Orientation Distribution Function (ODF) characterizes key brain microstructural properties and plays an important role in understanding brain structural connectivity. Recent works introduced Implicit Neural Representation (INR) based approaches to form a spatially aware continuous estimate of the ODF field and demonstrated promising results in key tasks of interest when compared to conventional discrete approaches. However, traditional INR methods face difficulties when scaling to large-scale images, such as modern ultra-high-resolution MRI scans, posing challenges in learning fine structures as well as inefficiencies in training and inference speed. In this work, we propose HashEnc, a grid-hash-encoding-based estimation of the ODF field and demonstrate its effectiveness in retaining structural and textural features. We show that HashEnc achieves a 10 % enhancement in image quality while requiring 3 $$\times $$ less computational resources than current methods. Our code can be found at https://github.com/MunzerDw/NODF-HashEnc .

Link prediction algorithms aim to infer the existence of connections (or links) between nodes in network-structured data and are typically applied to refine the connectivity among nodes. In this work, we focus on link prediction for flow-driven spatial networks, which are embedded in a Euclidean space and relate to physical exchange and transportation processes (e.g., blood flow in vessels or traffic flow in road networks). To this end, we propose the Graph Attentive Vectors (GAV) link prediction framework. GAV models simplified dynamics of physical flow in spatial networks via an attentive, neighborhood-aware message-passing paradigm, updating vector embeddings in a constrained manner. We evaluate GAV on eight flow-driven spatial networks given by whole-brain vessel graphs and road networks. GAV demonstrates superior performances across all datasets and metrics and outperformed the state-of-the-art on the ogbl-vessel benchmark at the time of submission by 12% (98.38 vs. 87.98 AUC). All code is publicly available on GitHub.1

Deep learning has potential to automate screening, monitoring and grading of disease in medical images. Pretraining with contrastive learning enables models to extract robust and generalisable features from natural image datasets, facilitating label-efficient downstream image analysis. However, the direct application of conventional contrastive methods to medical datasets introduces two domain-specific issues. Firstly, several image transformations which have been shown to be crucial for effective contrastive learning do not translate from the natural image to the medical image domain. Secondly, the assumption made by conventional methods, that any two images are dissimilar, is systematically misleading in medical datasets depicting the same anatomy and disease. This is exacerbated in longitudinal image datasets that repeatedly image the same patient cohort to monitor their disease progression over time. In this paper we tackle these issues by extending conventional contrastive frameworks with a novel metadata-enhanced strategy. Our approach employs widely available patient metadata to approximate the true set of inter-image contrastive relationships. To this end we employ records for patient identity, eye position (i.e. left or right) and time series information. In experiments using two large longitudinal datasets containing 170,427 retinal optical coherence tomography (OCT) images of 7912 patients with age-related macular degeneration (AMD), we evaluate the utility of using metadata to incorporate the temporal dynamics of disease progression into pretraining. Our metadata-enhanced approach outperforms both standard contrastive methods and a retinal image foundation model in five out of six image-level downstream tasks related to AMD. We find benefits in both a low-data and high-data regime across tasks ranging from AMD stage and type classification to prediction of visual acuity. Due to its modularity, our method can be quickly and cost-effectively tested to establish the potential benefits of including available metadata in contrastive pretraining.

The lack of reliable biomarkers makes predicting the conversion from intermediate to neovascular age-related macular degeneration (iAMD, nAMD) a challenging task. We develop a Deep Learning (DL) model to predict the future risk of conversion of an eye from iAMD to nAMD from its current OCT scan. Although eye clinics generate vast amounts of longitudinal OCT scans to monitor AMD progression, only a small subset can be manually labeled for supervised DL. To address this issue, we propose Morph-SSL, a novel Self-supervised Learning (SSL) method for longitudinal data. It uses pairs of unlabelled OCT scans from different visits and involves morphing the scan from the previous visit to the next. The Decoder predicts the transformation for morphing and ensures a smooth feature manifold that can generate intermediate scans between visits through linear interpolation. Next, the Morph-SSL trained features are input to a Classifier which is trained in a supervised manner to model the cumulative probability distribution of the time to conversion with a sigmoidal function. Morph-SSL was trained on unlabelled scans of 399 eyes (3570 visits). The Classifier was evaluated with a five-fold cross-validation on 2418 scans from 343 eyes with clinical labels of the conversion date. The Morph-SSL features achieved an AUC of 0.779 in predicting the conversion to nAMD within the next 6 months, outperforming the same network when trained end-to-end from scratch or pre-trained with popular SSL methods. Automated prediction of the future risk of nAMD onset can enable timely treatment and individualized AMD management.

Self-supervised learning (SSL) has emerged as a powerful technique for improving the efficiency and effectiveness of deep learning models. Contrastive methods are a prominent family of SSL that extract similar representations of two augmented views of an image while pushing away others in the representation space as negatives. However, the state-of-the-art contrastive methods require large batch sizes and augmentations designed for natural images that are impractical for 3D medical images. To address these limitations, we propose a new longitudinal SSL method, 3DTINC, based on non-contrastive learning. It is designed to learn perturbation-invariant features for 3D optical coherence tomography (OCT) volumes, using augmentations specifically designed for OCT. We introduce a new non-contrastive similarity loss term that learns temporal information implicitly from intra-patient scans acquired at different times. Our experiments show that this temporal information is crucial for predicting progression of retinal diseases, such as age-related macular degeneration (AMD). After pretraining with 3DTINC, we evaluated the learned representations and the prognostic models on two large-scale longitudinal datasets of retinal OCTs where we predict the conversion to wet-AMD within a six-month interval. Our results demonstrate that each component of our contributions is crucial for learning meaningful representations useful in predicting disease progression from longitudinal volumetric scans.

Accurate motion estimation at high acceleration factors enables rapid motion-compensated reconstruction in Magnetic Resonance Imaging (MRI) without compromising the diagnostic image quality. In this work, we introduce an attention-aware deep learning-based framework that can perform non-rigid pairwise registration for fully sampled and accelerated MRI. We extract local visual representations to build similarity maps between the registered image pairs at multiple resolution levels and additionally leverage long-range contextual information using a transformer-based module to alleviate ambiguities in the presence of artifacts caused by undersampling. We combine local and global dependencies to perform simultaneous coarse and fine motion estimation. The proposed method was evaluated on in-house acquired fully sampled and accelerated data of 101 patients and 62 healthy subjects undergoing cardiac and thoracic MRI. The impact of motion estimation accuracy on the downstream task of motion-compensated reconstruction was analyzed. We demonstrate that our model derives reliable and consistent motion fields across different sampling trajectories (Cartesian and radial) and acceleration factors of up to 16x for cardiac motion and 30x for respiratory motion and achieves superior image quality in motion-compensated reconstruction qualitatively and quantitatively compared to conventional and recent deep learning-based approaches. The code is publicly available at https://github.com/lab-midas/GMARAFT.

Ultra-wideband raster-scan optoacoustic mesoscopy (RSOM) is a novel modality that has demonstrated unprecedented ability to visualize epidermal and dermal structures in-vivo. However, an automatic and quantitative analysis of three-dimensional RSOM datasets remains unexplored. In this work we present our framework: Deep Learning RSOM Analysis Pipeline (DeepRAP), to analyze and quantify morphological skin features recorded by RSOM and extract imaging biomarkers for disease characterization. DeepRAP uses a multi-network segmentation strategy based on convolutional neural networks with transfer learning. This strategy enabled the automatic recognition of skin layers and subsequent segmentation of dermal microvasculature with an accuracy equivalent to human assessment. DeepRAP was validated against manual segmentation on 25 psoriasis patients under treatment and our biomarker extraction was shown to characterize disease severity and progression well with a strong correlation to physician evaluation and histology. In a unique validation experiment, we applied DeepRAP in a time series sequence of occlusion-induced hyperemia from 10 healthy volunteers. We observe how the biomarkers decrease and recover during the occlusion and release process, demonstrating accurate performance and reproducibility of DeepRAP. Furthermore, we analyzed a cohort of 75 volunteers and defined a relationship between aging and microvascular features in-vivo. More precisely, this study revealed that fine microvascular features in the dermal layer have the strongest correlation to age. The ability of our newly developed framework to enable the rapid study of human skin morphology and microvasculature in-vivo promises to replace biopsy studies, increasing the translational potential of RSOM.

Optical coherence tomography angiography (OCTA) is a non-invasive imaging modality that can acquire high-resolution volumes of the retinal vasculature and aid the diagnosis of ocular, neurological and cardiac diseases. Segmenting the visible blood vessels is a common first step when extracting quantitative biomarkers from these images. Classical segmentation algorithms based on thresholding are strongly affected by image artifacts and limited signal-to-noise ratio. The use of modern, deep learning-based segmentation methods has been inhibited by a lack of large datasets with detailed annotations of the blood vessels. To address this issue, recent work has employed transfer learning, where a segmentation network is trained on synthetic OCTA images and is then applied to real data. However, the previously proposed simulations fail to faithfully model the retinal vasculature and do not provide effective domain adaptation. Because of this, current methods are unable to fully segment the retinal vasculature, in particular the smallest capillaries. In this work, we present a lightweight simulation of the retinal vascular network based on space colonization for faster and more realistic OCTA synthesis. We then introduce three contrast adaptation pipelines to decrease the domain gap between real and artificial images. We demonstrate the superior segmentation performance of our approach in extensive quantitative and qualitative experiments on three public datasets that compare our method to traditional computer vision algorithms and supervised training using human annotations. Finally, we make our entire pipeline publicly available, including the source code, pretrained models, and a large dataset of synthetic OCTA images.

Deep unsupervised approaches are gathering increased attention for applications such as pathology detection and segmentation in medical images since they promise to alleviate the need for large labeled datasets and are more generalizable than their supervised counterparts in detecting any kind of rare pathology. As the Unsupervised Anomaly Detection (UAD) literature continuously grows and new paradigms emerge, it is vital to continuously evaluate and benchmark new methods in a common framework, in order to reassess the state-of-the-art (SOTA) and identify promising research directions. To this end, we evaluate a diverse selection of cutting-edge UAD methods on multiple medical datasets, comparing them against the established SOTA in UAD for brain MRI. Our experiments demonstrate that newly developed feature-modeling methods from the industrial and medical literature achieve increased performance compared to previous work and set the new SOTA in a variety of modalities and datasets. Additionally, we show that such methods are capable of benefiting from recently developed self-supervised pre-training algorithms, further increasing their performance. Finally, we perform a series of experiments in order to gain further insights into some unique characteristics of selected models and datasets. Our code can be found under https://github.com/iolag/UPD_study/.

The prevailing understanding in the field of machine learning and deep learning (ML/DL) is that, given a highquality dataset, one can effectively learn data-related priors through supervised learning. However, in medical imaging, this assumption faces two critical challenges: 1) high-quality training data are often scarce and 2) data are highly heterogeneous, stemming from different imaging scanners, protocols, or populations at various institutions. This diversity makes it impractical to represent the data with a single, universal prior using traditional methods, leading to limited generalizability in medical imaging tasks.

3D motion estimation from cine cardiac magnetic resonance (CMR) images is important for the assessment of cardiac function and the diagnosis of cardiovascular diseases. Current state-of-the art methods focus on estimating dense pixel-/voxel-wise motion fields in image space, which ignores the fact that motion estimation is only relevant and useful within the anatomical objects of interest, e.g., the heart. In this work, we model the heart as a 3D mesh consisting of epi- and endocardial surfaces. We propose a novel learning framework, DeepMesh, which propagates a template heart mesh to a subject space and estimates the 3D motion of the heart mesh from CMR images for individual subjects. In DeepMesh, the heart mesh of the end-diastolic frame of an individual subject is first reconstructed from the template mesh. Mesh-based 3D motion fields with respect to the end-diastolic frame are then estimated from 2D short- and long-axis CMR images. By developing a differentiable mesh-to-image rasterizer, DeepMesh is able to leverage 2D shape information from multiple anatomical views for 3D mesh reconstruction and mesh motion estimation. The proposed method estimates vertex-wise displacement and thus maintains vertex correspondences between time frames, which is important for the quantitative assessment of cardiac function across different subjects and populations. We evaluate DeepMesh on CMR images acquired from the UK Biobank. We focus on 3D motion estimation of the left ventricle in this work. Experimental results show that the proposed method quantitatively and qualitatively outperforms other image-based and mesh-based cardiac motion tracking methods.

In cardiac CINE, motion-compensated MR reconstruction (MCMR) is an effective approach to address highly undersampled acquisitions by incorporating motion information between frames. In this work, we propose a novel perspective for addressing the MCMR problem and a more integrated and efficient solution to the MCMR field. Contrary to state-of-the-art (SOTA) MCMR methods which break the original problem into two sub-optimization problems, i.e. motion estimation and reconstruction, we formulate this problem as a single entity with one single optimization. Our approach is unique in that the motion estimation is directly driven by the ultimate goal, reconstruction, but not by the canonical motion-warping loss (similarity measurement between motion-warped images and target images). We align the objectives of motion estimation and reconstruction, eliminating the drawbacks of artifacts-affected motion estimation and therefore error-propagated reconstruction. Further, we can deliver high-quality reconstruction and realistic motion without applying any regularization/smoothness loss terms, circumventing the non-trivial weighting factor tuning. We evaluate our method on two datasets: 1) an in-house acquired 2D CINE dataset for the retrospective study and 2) the public OCMR cardiac dataset for the prospective study. The conducted experiments indicate that the proposed MCMR framework can deliver artifact-free motion estimation and high-quality MR images even for imaging accelerations up to 20x, outperforming SOTA non-MCMR and MCMR methods in both qualitative and quantitative evaluation across all experiments. The code is available at https://github.com/JZPeterPan/MCMR-Recon-Driven-Motion.

Two key questions in cardiac image analysis are to assess the anatomy and motion of the heart from images; and to understand how they are associated with non-imaging clinical factors such as gender, age and diseases. While the first question can often be addressed by image segmentation and motion tracking algorithms, our capability to model and answer the second question is still limited. In this work, we propose a novel conditional generative model to describe the 4D spatio-temporal anatomy of the heart and its interaction with non-imaging clinical factors. The clinical factors are integrated as the conditions of the generative modelling, which allows us to investigate how these factors influence the cardiac anatomy. We evaluate the model performance in mainly two tasks, anatomical sequence completion and sequence generation. The model achieves high performance in anatomical sequence completion, comparable to or outperforming other state-of-the-art generative models. In terms of sequence generation, given clinical conditions, the model can generate realistic synthetic 4D sequential anatomies that share similar distributions with the real data. The code and the trained generative model are available at https://github.com/MengyunQ/CHeart.

Motion represents one of the major challenges in magnetic resonance imaging (MRI). Since the MR signal is acquired in frequency space, any motion of the imaged object leads to complex artefacts in the reconstructed image in addition to other MR imaging artefacts. Deep learning has been frequently proposed for motion correction at several stages of the reconstruction process. The wide range of MR acquisition sequences, anatomies and pathologies of interest, and motion patterns (rigid vs. deformable and random vs. regular) makes a comprehensive solution unlikely. To facilitate the transfer of ideas between different applications, this review provides a detailed overview of proposed methods for learning-based motion correction in MRI together with their common challenges and potentials. This review identifies differences and synergies in underlying data usage, architectures, training and evaluation strategies. We critically discuss general trends and outline future directions, with the aim to enhance interaction between different application areas and research fields.

Cardiovascular diseases are a significant cause of mortality worldwide, and the accurate diagnosis of these conditions is essential for effective treatment and management. Electrocardiograms (ECGs) are a common diagnostic tool used by cardiologists, but the manual interpretation of ECGs can be relatively time-consuming and challenging, particularly in cases of atrial fibrillation (AF), which is associated with an increased risk of stroke, heart failure, and other complications. To address the need for reliable and automatic ECG classifiers, we propose a new method using self-supervised learning with a CNNTransformer architecture to improve the ECG classification performance. The proposed model is pre-trained on the China Physiological Signal Challenge 2018 dataset and part of the Physikalisch-Technische Bundesanstalt (PTB) XL dataset using a novel ’nextclip’ prediction task, which asks the model to predict the next small segment of ECG, followed by finetuning on the ECG classification task. Our experimental results demonstrate that our proposed method achieves state-of-the-art results for ECG classification, with an average F1-score of 0.84 and 0.96 for AF detection on the CPSC2018 dataset. The proposed CNNTransformer architecture has shown to be an effective and efficient solution for ECG classification, especially on AF.

Medical datasets and especially biobanks, often contain extensive tabular data with rich clinical information in addition to images. In practice, clinicians typically have less data, both in terms of diversity and scale, but still wish to deploy deep learning solutions. Combined with increasing medical dataset sizes and expensive annotation costs, the necessity for unsupervised methods that can pretrain multimodally and predict unimodally has risen. To address these needs, we propose the first self-supervised contrastive learning framework that takes advantage of images and tabular data to train unimodal encoders. Our solution combines SimCLR and SCARF, two leading contrastive learning strategies, and is simple and effective. In our experiments, we demonstrate the strength of our framework by predicting risks of myocardial infarction and coronary artery disease (CAD) using cardiac MR images and 120 clinical features from 40,000 UK Biobank subjects. Furthermore, we show the generalizability of our approach to natural images using the DVM car advertisement dataset. We take advantage of the high interpretability of tabular data and through attribution and ablation experiments find that morphometric tabular features, describing size and shape, have outsized importance during the contrastive learning process and improve the quality of the learned embeddings. Finally, we introduce a novel form of supervised contrastive learning, label as a feature (LaaF), by appending the ground truth label as a tabular feature during multimodal pretraining, outperforming all supervised contrastive baselines.11https://github.com/paulhager/MMCL-Tabular-Imaging

The automatic generation of radiology reports has the potential to assist radiologists in the time-consuming task of report writing. Existing methods generate the full report from image-level features, failing to explicitly focus on anatomical regions in the image. We propose a simple yet effective region-guided report generation model that detects anatomical regions and then describes individual, salient regions to form the final report. While previous methods generate reports without the possibility of human intervention and with limited explainability, our method opens up novel clinical use cases through additional interactive capabilities and introduces a high degree of transparency and explainability. Comprehensive experiments demonstrate our method's effectiveness in report generation, outperforming previous state-of-the-art models, and highlight its interactive capabilities. The code and checkpoints are available at https://github.com/ttanida/rgrg.

The skeleton of a digital image is a compact representation of its topology, geometry, and scale. It has utility in many computer vision applications, such as image description, segmentation, and registration. However, skeletonization has only seen limited use in contemporary deep learning solutions. Most existing skeletonization algorithms are not differentiable, making it impossible to integrate them with gradient-based optimization. Compatible algorithms based on morphological operations and neural networks have been proposed, but their results often deviate from the geometry and topology of the true medial axis. This work introduces the first three-dimensional skeletonization algorithm that is both compatible with gradient-based optimization and preserves an object’s topology. Our method is exclusively based on matrix additions and multiplications, convolutional operations, basic non-linear functions, and sampling from a uniform probability distribution, allowing it to be easily implemented in any major deep learning library. In benchmarking experiments, we prove the advantages of our skeletonization algorithm compared to non-differentiable, morphological, and neural-network-based baselines. Finally, we demonstrate the utility of our algorithm by integrating it with two medical image processing applications that use gradient-based optimization: deep-learning-based blood vessel segmentation, and multimodal registration of the mandible in computed tomography and magnetic resonance images.

Differentially private mechanisms restrict the membership inference capabilities of powerful (optimal) adversaries against machine learning models. Such adversaries are rarely encountered in practice. In this work, we examine a more realistic threat model relaxation, where (sub-optimal) adversaries lack access to the exact model training database, but may possess related or partial data. We then formally characterise and experimentally validate adversarial membership inference capabilities in this setting in terms of hypothesis testing errors. Our work helps users to interpret the privacy properties of sensitive data processing systems under realistic threat model relaxations and choose appropriate noise levels for their use-case.

Myocardial motion and deformation are rich descriptors that characterize cardiac function. Image registration, as the most commonly used technique for myocardial motion tracking, is an ill-posed inverse problem which often requires prior assumptions on the solution space. In contrast to most existing approaches which impose explicit generic regularization such as smoothness, in this work we propose a novel method that can implicitly learn an application-specific biomechanics-informed prior and embed it into a neural network-parameterized transformation model. Particularly, the proposed method leverages a variational autoencoder-based generative model to learn a manifold for biomechanically plausible deformations. The motion tracking then can be performed via traversing the learnt manifold to search for the optimal transformations while considering the sequence information. The proposed method is validated on three public cardiac cine MRI datasets with comprehensive evaluations. The results demonstrate that the proposed method can outperform other approaches, yielding higher motion tracking accuracy with reasonable volume preservation and better generalizability to varying data distributions. It also enables better estimates of myocardial strains, which indicates the potential of the method in characterizing spatiotemporal signatures for understanding cardiovascular diseases.

Automatic segmentation of the placenta in fetal ultrasound (US) is challenging due to the (i) high diversity of placenta appearance, (ii) the restricted quality in US resulting in highly variable reference annotations, and (iii) the limited field-of-view of US prohibiting whole placenta assessment at late gestation. In this work, we address these three challenges with a multi-task learning approach that combines the classification of placental location (e.g., anterior, posterior) and semantic placenta segmentation in a single convolutional neural network. Through the classification task the model can learn from larger and more diverse datasets while improving the accuracy of the segmentation task in particular in limited training set conditions. With this approach we investigate the variability in annotations from multiple raters and show that our automatic segmentations (Dice of 0.86 for anterior and 0.83 for posterior placentas) achieve human-level performance as compared to intra- and inter-observer variability. Lastly, our approach can deliver whole placenta segmentation using a multi-view US acquisition pipeline consisting of three stages: multi-probe image acquisition, image fusion and image segmentation. This results in high quality segmentation of larger structures such as the placenta in US with reduced image artifacts which are beyond the field-of-view of single probes.

Graph Neural Networks (GNNs) have established themselves as state-of-the-art for many machine learning applications such as the analysis of social and medical networks. Several among these datasets contain privacy-sensitive data. Machine learning with differential privacy is a promising technique to allow deriving insight from sensitive data while offering formal guarantees of privacy protection. However, the differentially private training of GNNs has so far remained under-explored due to the challenges presented by the intrinsic structural connectivity of graphs. In this work, we introduce a framework for differential private graph-level classification. Our method is applicable to graph deep learning on multi-graph datasets and relies on differentially private stochastic gradient descent (DP-SGD). We show results on a variety of datasets and evaluate the impact of different GNN architectures and training hyperparameters on model performance for differentially private graph classification, as well as the scalability of the method on a large medical dataset. Our experiments show that DP-SGD can be applied to graph classification tasks with reasonable utility losses. Furthermore, we apply explainability techniques to assess whether similar representations are learned in the private and non-private settings. Our results can also function as robust baselines for future work in this area.

Collaborative machine learning settings such as federated learning can be susceptible to adversarial interference and attacks. One class of such attacks is termed model inversion attacks , characterised by the adversary reverse-engineering the model into disclosing the training data. Previous implementations of this attack typically only rely on the shared data representations, ignoring the adversarial priors, or require that specific layers are present in the target model, reducing the potential attack surface. In this work, we propose a novel context-agnostic model inversion framework that builds on the foundations of gradient-based inversion attacks, but additionally exploits the features and the style of the data controlled by an in-the-network adversary. Our technique outperforms existing gradient-based approaches both qualitatively and quantitatively across all training settings, showing particular effectiveness against the collaborative medical imaging tasks. Finally, we demonstrate that our method achieves significant success on two downstream tasks: sensitive feature inference and facial recognition spoofing.

We present CortexODE, a deep learning framework for cortical surface reconstruction. CortexODE leverages neural ordinary differential equations (ODEs) to deform an input surface into a target shape by learning a diffeomorphic flow. The trajectories of the points on the surface are modeled as ODEs, where the derivatives of their coordinates are parameterized via a learnable Lipschitz-continuous deformation network. This provides theoretical guarantees for the prevention of self-intersections. CortexODE can be integrated to an automatic learning-based pipeline, which reconstructs cortical surfaces efficiently in less than 5 seconds. The pipeline utilizes a 3D U-Net to predict a white matter segmentation from brain Magnetic Resonance Imaging (MRI) scans, and further generates a signed distance function that represents an initial surface. Fast topology correction is introduced to guarantee homeomorphism to a sphere. Following the isosurface extraction step, two CortexODE models are trained to deform the initial surface to white matter and pial surfaces respectively. The proposed pipeline is evaluated on large-scale neuroimage datasets in various age groups including neonates (25-45 weeks), young adults (22-36 years) and elderly subjects (55-90 years). Our experiments demonstrate that the CortexODE-based pipeline can achieve less than 0.2mm average geometric error while being orders of magnitude faster compared to conventional processing pipelines.

The cornerstone of stroke care is expedient management that varies depending on the time since stroke onset. Consequently, clinical decision making is centered on accurate knowledge of timing and often requires a radiologist to interpret Computed Tomography (CT) of the brain to confirm the occurrence and age of an event. These tasks are particularly challenging due to the subtle expression of acute ischemic lesions and the dynamic nature of their appearance. Automation efforts have not yet applied deep learning to estimate lesion age and treated these two tasks independently, so, have overlooked their inherent complementary relationship. To leverage this, we propose a novel end-to-end multi-task transformer-based network optimized for concurrent segmentation and age estimation of cerebral ischemic lesions. By utilizing gated positional self-attention and CT-specific data augmentation, the proposed method can capture long-range spatial dependencies while maintaining its ability to be trained from scratch under low-data regimes commonly found in medical imaging. Furthermore, to better combine multiple predictions, we incorporate uncertainty by utilizing quantile loss to facilitate estimating a probability density function of lesion age. The effectiveness of our model is then extensively evaluated on a clinical dataset consisting of 776 CT images from two medical centers. Experimental results demonstrate that our method obtains promising performance, with an area under the curve (AUC) of 0.933 for classifying lesion ages ≤ 4.5 hours compared to 0.858 using a conventional approach, and outperforms task-specific state-of-the-art algorithms.

Deep learning models usually suffer from the domain shift issue, where models trained on one source domain do not generalize well to other unseen domains. In this work, we investigate the single-source domain generalization problem: training a deep network that is robust to unseen domains, under the condition that training data are only available from one source domain, which is common in medical imaging applications. We tackle this problem in the context of cross-domain medical image segmentation. In this scenario, domain shifts are mainly caused by different acquisition processes. We propose a simple causality-inspired data augmentation approach to expose a segmentation model to synthesized domain-shifted training examples. Specifically, 1) to make the deep model robust to discrepancies in image intensities and textures, we employ a family of randomly-weighted shallow networks. They augment training images using diverse appearance transformations. 2) Further we show that spurious correlations among objects in an image are detrimental to domain robustness. These correlations might be taken by the network as domain-specific clues for making predictions, and they may break on unseen domains. We remove these spurious correlations via causal intervention. This is achieved by resampling the appearances of potentially correlated objects independently. The proposed approach is validated on three cross-domain segmentation scenarios: cross-modality (CT-MRI) abdominal image segmentation, cross-sequence (bSSFP-LGE) cardiac MRI segmentation, and cross-site prostate MRI segmentation. The proposed approach yields consistent performance gains compared with competitive methods when tested on unseen domains.

Contrastive learning has proven effective for pre-training image models on unlabeled data with promising results for tasks such as medical image classification. Using paired text (like radiological reports) during pre-training improves the results even further. Still, most existing methods target image classification downstream tasks and may not be optimal for localized tasks like semantic segmentation or object detection. We therefore propose Localized representation learning from Vision and Text (LoVT), to our best knowledge, the first text-supervised pre-training method that targets localized medical imaging tasks. Our method combines instance-level image-report contrastive learning with local contrastive learning on image region and report sentence representations. We evaluate LoVT and commonly used pre-training methods on an evaluation framework of 18 localized tasks on chest X-rays from five public datasets. LoVT performs best on 10 of the 18 studied tasks making it the preferred method of choice for localized tasks.

The success of neural networks on medical image segmentation tasks typically relies on large labeled datasets for model training. However, acquiring and manually labeling a large medical image set is resource-intensive, expensive, and sometimes impractical due to data sharing and privacy issues. To address this challenge, we propose AdvChain, a generic adversarial data augmentation framework, aiming at improving both the diversity and effectiveness of training data for medical image segmentation tasks. AdvChain augments data with dynamic data augmentation, generating randomly chained photo-metric and geometric transformations to resemble realistic yet challenging imaging variations to expand training data. By jointly optimizing the data augmentation model and a segmentation network during training, challenging examples are generated to enhance network generalizability for the downstream task. The proposed adversarial data augmentation does not rely on generative networks and can be used as a plug-in module in general segmentation networks. It is computationally efficient and applicable for both low-shot supervised and semi-supervised learning. We analyze and evaluate the method on two MR image segmentation tasks: cardiac segmentation and prostate segmentation with limited labeled data. Results show that the proposed approach can alleviate the need for labeled data while improving model generalization ability, indicating its practical value in medical imaging applications.

Compressed sensing (CS) has been playing a key role in accelerating the magnetic resonance imaging (MRI) acquisition process. With the resurgence of artificial intelligence, deep neural networks and CS algorithms are being integrated to redefine the state of the art of fast MRI. The past several years have witnessed substantial growth in the complexity, diversity, and performance of deep-learning-based CS techniques that are dedicated to fast MRI. In this meta-analysis, we systematically review the deep-learning-based CS techniques for fast MRI, describe key model designs, highlight breakthroughs, and discuss promising directions. We have also introduced a comprehensive analysis framework and a classification system to assess the pivotal role of deep learning in CS-based acceleration for MRI.

Abstract In this study, we aim to bridge the gap between the theoretical understanding of attacks against collaborative machine learning workflows and their practical ramifications by considering the effects of model architecture, learning setting and hyperparameters on the resilience against attacks. We refer to such mitigations asmodel adaptation. Through extensive experimentation on both, benchmark and real-life datasets, we establish a more practical threat model for collaborative learning scenarios. In particular, we evaluate the impact of model adaptation by implementing a range of attacks belonging to the broader categories of model inversion and membership inference. Our experiments yield two noteworthy outcomes: they demonstrate the difficulty of actually conducting successful attacks under realistic settings when model adaptation is employed and they highlight the challenge inherent in successfully combining model adaptation and formal privacy-preserving techniques to retain the optimal balance between model utility and attack resilience.

Intelligent video summarization algorithms allow to quickly convey the most relevant information in videos through the identification of the most essential and explanatory content while removing redundant video frames. In this paper, we introduce the 3DST-UNet-RL framework for video summarization. A 3D spatio-temporal U-Net is used to efficiently encode spatio-temporal information of the input videos for downstream reinforcement learning (RL). An RL agent learns from spatio-temporal latent scores and predicts actions for keeping or rejecting a video frame in a video summary. We investigate if real/inflated 3D spatio-temporal CNN features are better suited to learn representations from videos than commonly used 2D image features. Our framework can operate in both, a fully unsupervised mode and a supervised training mode. We analyse the impact of prescribed summary lengths and show experimental evidence for the effectiveness of 3DST-UNet-RL on two commonly used general video summarization benchmarks. We also applied our method on a medical video summarization task. The proposed video summarization method has the potential to save storage costs of ultrasound screening videos as well as to increase efficiency when browsing patient video data during retrospective analysis or audit without loosing essential information.

Data-driven deep learning approaches to image registration can be less accurate than conventional iterative approaches, especially when training data is limited. To address this issue and meanwhile retain the fast inference speed of deep learning, we propose VR-Net, a novel cascaded variational network for unsupervised deformable image registration. Using a variable splitting optimization scheme, we first convert the image registration problem, established in a generic variational framework, into two sub-problems, one with a point-wise, closed-form solution and the other one being a denoising problem. We then propose two neural layers (i.e. warping layer and intensity consistency layer) to model the analytical solution and a residual U-Net (termed generalized denoising layer) to formulate the denoising problem. Finally, we cascade the three neural layers multiple times to form our VR-Net. Extensive experiments on three (two 2D and one 3D) cardiac magnetic resonance imaging datasets show that VR-Net outperforms state-of-the-art deep learning methods on registration accuracy, whilst maintaining the fast inference speed of deep learning and the data-efficiency of variational models.

Recovering the 3D motion of the heart from cine cardiac magnetic resonance (CMR) imaging enables the assessment of regional myocardial function and is important for understanding and analyzing cardiovascular disease. However, 3D cardiac motion estimation is challenging because the acquired cine CMR images are usually 2D slices which limit the accurate estimation of through-plane motion. To address this problem, we propose a novel multi-view motion estimation network (MulViMotion), which integrates 2D cine CMR images acquired in short-axis and long-axis planes to learn a consistent 3D motion field of the heart. In the proposed method, a hybrid 2D/3D network is built to generate dense 3D motion fields by learning fused representations from multi-view images. To ensure that the motion estimation is consistent in 3D, a shape regularization module is introduced during training, where shape information from multi-view images is exploited to provide weak supervision to 3D motion estimation. We extensively evaluate the proposed method on 2D cine CMR images from 580 subjects of the UK Biobank study for 3D motion tracking of the left ventricular myocardium. Experimental results show that the proposed method quantitatively and qualitatively outperforms competing methods.

Fully-supervised deep learning segmentation models are inflexible when encountering new unseen semantic classes and their fine-tuning often requires significant amounts of annotated data. Few-shot semantic segmentation (FSS) aims to solve this inflexibility by learning to segment an arbitrary unseen semantically meaningful class by referring to only a few labeled examples, without involving fine-tuning. State-of-the-art FSS methods are typically designed for segmenting natural images and rely on abundant annotated data of training classes to learn image representations that generalize well to unseen testing classes. However, such a training mechanism is impractical in annotation-scarce medical imaging scenarios. To address this challenge, in this work, we propose a novel self-supervised FSS framework for medical images, named SSL-ALPNet, in order to bypass the requirement for annotations during training. The proposed method exploits superpixel-based pseudo-labels to provide supervision signals. In addition, we propose a simple yet effective adaptive local prototype pooling module which is plugged into the prototype networks to further boost segmentation accuracy. We demonstrate the general applicability of the proposed approach using three different tasks: organ segmentation of abdominal CT and MRI images respectively, and cardiac segmentation of MRI images. The proposed method yields higher Dice scores than conventional FSS methods which require manual annotations for training in our experiments.

Since its renaissance, deep learning has been widely used in various medical imaging tasks and has achieved remarkable success in many medical imaging applications, thereby propelling us into the so-called artificial intelligence (AI) era. It is known that the success of AI is mostly attributed to the availability of big data with annotations for a single task and the advances in high performance computing. However, medical imaging presents unique challenges that confront deep learning approaches. In this survey paper, we first present traits of medical imaging, highlight both clinical needs and technical challenges in medical imaging, and describe how emerging trends in deep learning are addressing these issues. We cover the topics of network architecture, sparse and noisy labels, federating learning, interpretability, uncertainty quantification, etc. Then, we present several case studies that are commonly found in clinical practice, including digital pathology and chest, brain, cardiovascular, and abdominal imaging. Rather than presenting an exhaustive literature survey, we instead describe some prominent research highlights related to these case study applications. We conclude with a discussion and presentation of promising future directions.

Deep neural networks exhibit limited generalizability across images with different entangled domain features and categorical features. Learning generalizable features that can form universal categorical decision boundaries across domains is an interesting and difficult challenge. This problem occurs frequently in medical imaging applications when attempts are made to deploy and improve deep learning models across different image acquisition devices, across acquisition parameters or if some classes are unavailable in new training databases. To address this problem, we propose Mutual Information-based Disentangled Neural Networks (MIDNet), which extract generalizable categorical features to transfer knowledge to unseen categories in a target domain. The proposed MIDNet adopts a semi-supervised learning paradigm to alleviate the dependency on labeled data. This is important for real-world applications where data annotation is time-consuming, costly and requires training and expertise. We extensively evaluate the proposed method on fetal ultrasound datasets for two different image classification tasks where domain features are respectively defined by shadow artifacts and image acquisition devices. Experimental results show that the proposed method outperforms the state-of-the-art on the classification of unseen categories in a target domain with sparsely labeled training data.

Model-based approaches for image reconstruction, analysis, and interpretation have made significant progress over the past decades. Many of these approaches are based on either mathematical, physical, or biological models. A challenge for these approaches is the modeling of the underlying processes (e.g., the physics of image acquisition or the patho-physiology of a disease) with appropriate levels of detail and realism. With the availability of large amounts of imaging data and machine learning (in particular deep learning) techniques, data-driven approaches have become more widespread for use in different tasks in reconstruction, analysis, and interpretation. These approaches learn statistical models directly from labeled or unlabeled image data and have been shown to be very powerful for extracting clinically useful information from medical imaging. While these data-driven approaches often outperform traditional model-based approaches, their clinical deployment often poses challenges in terms of robustness, generalization ability, and interpretability. In this article, we discuss what developments have motivated the shift from model-based approaches toward data-driven strategies and what potential problems are associated with the move toward purely data-driven approaches, in particular deep learning. We also discuss some of the open challenges for data-driven approaches, e.g., generalization to new unseen data (e.g., transfer learning), robustness to adversarial attacks, and interpretability. Finally, we conclude with a discussion on how these approaches may lead to the development of more closely coupled imaging pipelines that are optimized in an end-to-end fashion.

Quantification of anatomical shape changes currently relies on scalar global indexes which are largely insensitive to regional or asymmetric modifications. Accurate assessment of pathology-driven anatomical remodeling is a crucial step for the diagnosis and treatment of many conditions. Deep learning approaches have recently achieved wide success in the analysis of medical images, but they lack interpretability in the feature extraction and decision processes. In this work, we propose a new interpretable deep learning model for shape analysis. In particular, we exploit deep generative networks to model a population of anatomical segmentations through a hierarchy of conditional latent variables. At the highest level of this hierarchy, a two-dimensional latent space is simultaneously optimised to discriminate distinct clinical conditions, enabling the direct visualisation of the classification space. Moreover, the anatomical variability encoded by this discriminative latent space can be visualised in the segmentation space thanks to the generative properties of the model, making the classification task transparent. This approach yielded high accuracy in the categorisation of healthy and remodelled left ventricles when tested on unseen segmentations from our own multi-centre dataset as well as in an external validation set, and on hippocampi from healthy controls and patients with Alzheimer's disease when tested on ADNI data. More importantly, it enabled the visualisation in three-dimensions of both global and regional anatomical features which better discriminate between the conditions under exam. The proposed approach scales effectively to large populations, facilitating high-throughput analysis of normal anatomy and pathology in large-scale studies of volumetric imaging.

A number of studies have been developed recently in order to explore associations between craniofacial differences and genetics. Most of these works have been based on spatial face image models, adjusted for the counter effects of age. This approach provides a limited understanding of normal and abnormal craniofacial development owing to the lack of age progression information. Here, the authors propose and implement an imaging framework that combines facial landmark positioning, non‐rigid registration, novel age‐dependent face modelling and common distance metrics to disclose the most facial differences that vary across the time due to the subjects' age. All the experiments carried out and corresponding results presented here are based on a database comprising ordinary two‐dimensional (2D) frontal face images of Down Syndrome (DS) and control sample groups. A number of craniofacial metrics have been successfully identified that highlight statistically significant and clinically relevant differences between the controls and the faces associated with DS within the age range from 1 to 18 years old, producing realistic unbiased face models with similar level of detail at all age‐intervals, despite the small sample size available.

Good quality of medical images is a prerequisite for the success of subsequent image analysis pipelines. Quality assessment of medical images is therefore an essential activity and for large population studies such as the UK Biobank (UKBB), manual identification of artefacts such as those caused by unanticipated motion is tedious and time-consuming. Therefore, there is an urgent need for automatic image quality assessment techniques. In this paper, we propose a method to automatically detect the presence of motion-related artefacts in cardiac magnetic resonance (CMR) cine images. We compare two deep learning architectures to classify poor quality CMR images: 1) 3D spatio-temporal Convolutional Neural Networks (3D-CNN), 2) Long-term Recurrent Convolutional Network (LRCN). Though in real clinical setup motion artefacts are common, high-quality imaging of UKBB, which comprises cross-sectional population data of volunteers who do not necessarily have health problems creates a highly imbalanced classification problem. Due to the high number of good quality images compared to the relatively low number of images with motion artefacts, we propose a novel data augmentation scheme based on synthetic artefact creation in k-space. We also investigate a learning approach using a predetermined curriculum based on synthetic artefact severity. We evaluate our pipeline on a subset of the UK Biobank data set consisting of 3510 CMR images. The LRCN architecture outperformed the 3D-CNN architecture and was able to detect 2D+time short axis images with motion artefacts in less than 1ms with high recall. We compare our approach to a range of state-of-the-art quality assessment methods. The novel data augmentation and curriculum learning approaches both improved classification performance achieving overall area under the ROC curve of 0.89.

We propose a novel attention gate (AG) model for medical image analysis that automatically learns to focus on target structures of varying shapes and sizes. Models trained with AGs implicitly learn to suppress irrelevant regions in an input image while highlighting salient features useful for a specific task. This enables us to eliminate the necessity of using explicit external tissue/organ localisation modules when using convolutional neural networks (CNNs). AGs can be easily integrated into standard CNN models such as VGG or U-Net architectures with minimal computational overhead while increasing the model sensitivity and prediction accuracy. The proposed AG models are evaluated on a variety of tasks, including medical image classification and segmentation. For classification, we demonstrate the use case of AGs in scan plane detection for fetal ultrasound screening. We show that the proposed attention mechanism can provide efficient object localisation while improving the overall prediction performance by reducing false positives. For segmentation, the proposed architecture is evaluated on two large 3D CT abdominal datasets with manual annotations for multiple organs. Experimental results show that AG models consistently improve the prediction performance of the base architectures across different datasets and training sizes while preserving computational efficiency. Moreover, AGs guide the model activations to be focused around salient regions, which provides better insights into how model predictions are made. The source code for the proposed AG models is publicly available.

This paper presents a morphological analysis of fibrotic scarring in non-ischemic dilated cardiomyopathy, and its relationship to electrical instabilities which underlie reentrant arrhythmias.Two dimensional electrophysiological simulation models were constructed from a set of 699 late gadolinium enhanced cardiac magnetic resonance images originating from 157 patients.Areas of late gadolinium enhancement (LGE) in each image were assigned one of 10 possible microstructures, which modelled the details of fibrotic scarring an order of magnitude below the MRI scan resolution.A simulated programmed electrical stimulation protocol tested each model for the possibility of generating either a transmural block or a transmural reentry.The outcomes of the simulations were compared against morphological LGE features extracted from the images.Models which blocked or reentered, grouped by microstructure, were significantly different from one another in myocardial-LGE interface length, number of components and entropy, but not in relative area and transmurality.With an unknown microstructure, transmurality alone was the best predictor of block, whereas a combination of interface length, transmurality and number of components was the best predictor of reentry in linear discriminant analysis. Author summaryNon-ischemic dilated cardiomyopathy is a disease in which the lower left chamber of the heart is abnormally large.The cause of the disease can be anything that is not a loss of blood supply to the heart.Many patients with non-ischemic dilated cardiomyopathy have scars in their hearts which can be detected with magnetic resonance imaging.These scars are thought to disrupt the flow of electricity through the heart and cause deadly rhythm

Deep learning approaches have achieved state-of-the-art performance in cardiac magnetic resonance (CMR) image segmentation. However, most approaches have focused on learning image intensity features for segmentation, whereas the incorporation of anatomical shape priors has received less attention. In this paper, we combine a multi-task deep learning approach with atlas propagation to develop a shape-refined bi-ventricular segmentation pipeline for short-axis CMR volumetric images. The pipeline first employs a fully convolutional network (FCN) that learns segmentation and landmark localization tasks simultaneously. The architecture of the proposed FCN uses a 2.5D representation, thus combining the computational advantage of 2D FCNs networks and the capability of addressing 3D spatial consistency without compromising segmentation accuracy. Moreover, a refinement step is designed to explicitly impose shape prior knowledge and improve segmentation quality. This step is effective for overcoming image artifacts (e.g., due to different breath-hold positions and large slice thickness), which preclude the creation of anatomically meaningful 3D cardiac shapes. The pipeline is fully automated, due to network's ability to infer landmarks, which are then used downstream in the pipeline to initialize atlas propagation. We validate the pipeline on 1831 healthy subjects and 649 subjects with pulmonary hypertension. Extensive numerical experiments on the two datasets demonstrate that our proposed method is robust and capable of producing accurate, high-resolution, and anatomically smooth bi-ventricular 3D models, despite the presence of artifacts in input CMR volumes.

Detecting acoustic shadows in ultrasound images is important in many clinical and engineering applications. Real-time feedback of acoustic shadows can guide sonographers to a standardized diagnostic viewing plane with minimal artifacts and can provide additional information for other automatic image analysis algorithms. However, automatically detecting shadow regions using learning-based algorithms is challenging because pixel-wise ground truth annotation of acoustic shadows is subjective and time consuming. In this paper, we propose a weakly supervised method for automatic confidence estimation of acoustic shadow regions. Our method is able to generate a dense shadow-focused confidence map. In our method, a shadow-seg module is built to learn general shadow features for shadow segmentation, based on global image-level annotations as well as a small number of coarse pixel-wise shadow annotations. A transfer function is introduced to extend the obtained binary shadow segmentation to a reference confidence map. In addition, a confidence estimation network is proposed to learn the mapping between input images and the reference confidence maps. This network is able to predict shadow confidence maps directly from input images during inference. We use evaluation metrics such as DICE, inter-class correlation, and so on, to verify the effectiveness of our method. Our method is more consistent than human annotation and outperforms the state-of-the-art quantitatively in shadow segmentation and qualitatively in confidence estimation of shadow regions. Furthermore, we demonstrate the applicability of our method by integrating shadow confidence maps into tasks such as ultrasound image classification, multi-view image fusion, and automated biometric measurements.

Accelerating the data acquisition of dynamic magnetic resonance imaging leads to a challenging ill-posed inverse problem, which has received great interest from both the signal processing and machine learning communities over the last decades. The key ingredient to the problem is how to exploit the temporal correlations of the MR sequence to resolve aliasing artifacts. Traditionally, such observation led to a formulation of an optimization problem, which was solved using iterative algorithms. Recently, however, deep learning-based approaches have gained significant popularity due to their ability to solve general inverse problems. In this paper, we propose a unique, novel convolutional recurrent neural network architecture which reconstructs high quality cardiac MR images from highly undersampled k-space data by jointly exploiting the dependencies of the temporal sequences as well as the iterative nature of the traditional optimization algorithms. In particular, the proposed architecture embeds the structure of the traditional iterative algorithms, efficiently modeling the recurrence of the iterative reconstruction stages by using recurrent hidden connections over such iterations. In addition, spatio-temporal dependencies are simultaneously learnt by exploiting bidirectional recurrent hidden connections across time sequences. The proposed method is able to learn both the temporal dependence and the iterative reconstruction process effectively with only a very small number of parameters, while outperforming current MR reconstruction methods in terms of reconstruction accuracy and speed.

The effectiveness of a cardiovascular magnetic resonance (CMR) scan depends on the ability of the operator to correctly tune the acquisition parameters to the subject being scanned and on the potential occurrence of imaging artifacts, such as cardiac and respiratory motion. In the clinical practice, a quality control step is performed by visual assessment of the acquired images; however, this procedure is strongly operator-dependent, cumbersome, and sometimes incompatible with the time constraints in clinical settings and large-scale studies. We propose a fast, fully automated, and learning-based quality control pipeline for CMR images, specifically for short-axis image stacks. Our pipeline performs three important quality checks: 1) heart coverage estimation; 2) inter-slice motion detection; 3) image contrast estimation in the cardiac region. The pipeline uses a hybrid decision forest method-integrating both regression and structured classification models-to extract landmarks and probabilistic segmentation maps from both long- and short-axis images as a basis to perform the quality checks. The technique was tested on up to 3000 cases from the UK Biobank and on 100 cases from the UK Digital Heart Project and validated against manual annotations and visual inspections performed by expert interpreters. The results show the capability of the proposed pipeline to correctly detect incomplete or corrupted scans (e.g., on UK Biobank, sensitivity and specificity, respectively, 88% and 99% for heart coverage estimation and 85% and 95% for motion detection), allowing their exclusion from the analyzed dataset or the triggering of a new acquisition.

We present a novel cost function for semi-supervised learning of neural networks that encourages compact clustering of the latent space to facilitate separation. The key idea is to dynamically create a graph over embeddings of labeled and unlabeled samples of a training batch to capture underlying structure in feature space, and use label propagation to estimate its high and low density regions. We then devise a cost function based on Markov chains on the graph that regularizes the latent space to form a single compact cluster per class, while avoiding to disturb existing clusters during optimization. We evaluate our approach on three benchmarks and compare to state-of-the art with promising results. Our approach combines the benefits of graph-based regularization with efficient, inductive inference, does not require modifications to a network architecture, and can thus be easily applied to existing networks to enable an effective use of unlabeled data.

Convolutional neural networks have been widely used in medical image segmentation. The amount of training data strongly determines the overall performance. Most approaches are applied for a single imaging modality, e.g., brain MRI. In practice, it is often difficult to acquire sufficient training data of a certain imaging modality. The same anatomical structures, however, may be visible in different modalities such as major organs on abdominal CT and MRI. In this work, we investigate the effectiveness of learning from multiple modalities to improve the segmentation accuracy on each individual modality. We study the feasibility of using a dual-stream encoder-decoder architecture to learn modality-independent, and thus, generalisable and robust features. All of our MRI and CT data are unpaired, which means they are obtained from different subjects and not registered to each other. Experiments show that multi-modal learning can improve overall accuracy over modality-specific training. Results demonstrate that information across modalities can in particular improve performance on varying structures such as the spleen.

Motivation: Left ventricular (LV) hypertrophy is a strong predictor of cardiovascular outcomes, but its genetic regulation remains largely unexplained. Conventional phenotyping relies on manual calculation of LV mass and wall thickness, but advanced cardiac image analysis presents an opportunity for high-throughput mapping of genotype-phenotype associations in three dimensions (3D). Results: High-resolution cardiac magnetic resonance images were automatically segmented in 1124 healthy volunteers to create a 3D shape model of the heart. Mass univariate regression was used to plot a 3D effect-size map for the association between wall thickness and a set of predictors at each vertex in the mesh. The vertices where a significant effect exists were determined by applying threshold-free cluster enhancement to boost areas of signal with spatial contiguity. Experiments on simulated phenotypic signals and SNP replication show that this approach offers a substantial gain in statistical power for cardiac genotype-phenotype associations while providing good control of the false discovery rate. This framework models the effects of genetic variation throughout the heart and can be automatically applied to large population cohorts. Availability and implementation: The proposed approach has been coded in an R package freely available at https://doi.org/10.5281/zenodo.834610 together with the clinical data used in this work. Contact: [email protected]. Supplementary information: Supplementary data are available at Bioinformatics online.

Abnormal cardiac motion can indicate different forms of disease, which can manifest at different spatial scales in the myocardium. Many studies have sought to characterise particular motion abnormalities associated with specific diseases, and to utilise motion information to improve diagnoses. However, the importance of spatial scale in the analysis of cardiac deformation has not been extensively investigated. We build on recent work on the analysis of myocardial strains at different spatial scales using a cardiac motion atlas to find the optimal scales for estimating different cardiac biomarkers. We apply a multi-scale strain analysis to a 43 patient cohort of cardiac resynchronisation therapy (CRT) patients using tagged magnetic resonance imaging data for (1) predicting response to CRT, (2) identifying septal flash, (3) estimating QRS duration, and (4) identifying the presence of ischaemia. A repeated, stratified cross-validation is used to demonstrate the importance of spatial scale in our analysis, revealing different optimal spatial scales for the estimation of different biomarkers.

Multi-atlas segmentation is a widely used tool in medical image analysis, providing robust and accurate results by learning from annotated atlas datasets. However, the availability of fully annotated atlas images for training is limited due to the time required for the labelling task. Segmentation methods requiring only a proportion of each atlas image to be labelled could therefore reduce the workload on expert raters tasked with annotating atlas images. To address this issue, we first re-examine the labelling problem common in many existing approaches and formulate its solution in terms of a Markov Random Field energy minimisation problem on a graph connecting atlases and the target image. This provides a unifying framework for multi-atlas segmentation. We then show how modifications in the graph configuration of the proposed framework enable the use of partially annotated atlas images and investigate different partial annotation strategies. The proposed method was evaluated on two Magnetic Resonance Imaging (MRI) datasets for hippocampal and cardiac segmentation. Experiments were performed aimed at (1) recreating existing segmentation techniques with the proposed framework and (2) demonstrating the potential of employing sparsely annotated atlas data for multi-atlas segmentation.

Statistical shape modeling is a powerful tool for visualizing and quantifying geometric and functional patterns of the heart. After myocardial infarction (MI), the left ventricle typically remodels in response to physiological challenges. Several methods have been proposed in the literature to describe statistical shape changes. Which method best characterizes left ventricular remodeling after MI is an open research question. A better descriptor of remodeling is expected to provide a more accurate evaluation of disease status in MI patients. We therefore designed a challenge to test shape characterization in MI given a set of three-dimensional left ventricular surface points. The training set comprised 100 MI patients, and 100 asymptomatic volunteers (AV). The challenge was initiated in 2015 at the Statistical Atlases and Computational Models of the Heart workshop, in conjunction with the MICCAI conference. The training set with labels was provided to participants, who were asked to submit the likelihood of MI from a different (validation) set of 200 cases (100 AV and 100 MI). Sensitivity, specificity, accuracy and area under the receiver operating characteristic curve were used as the outcome measures. The goals of this challenge were to (1) establish a common dataset for evaluating statistical shape modeling algorithms in MI, and (2) test whether statistical shape modeling provides additional information characterizing MI patients over standard clinical measures. Eleven groups with a wide variety of classification and feature extraction approaches participated in this challenge. All methods achieved excellent classification results with accuracy ranges from 0.83 to 0.98. The areas under the receiver operating characteristic curves were all above 0.90. Four methods showed significantly higher performance than standard clinical measures. The dataset and software for evaluation are available from the Cardiac Atlas Project website1.

Convolutional neural networks (CNNs) have revolutionized medical image analysis over the past few years. The U-Net architecture is one of the most well-known CNN architectures for semantic segmentation and has achieved remarkable successes in many different medical image segmentation applications. The U-Net architecture consists of standard convolution layers, pooling layers, and upsampling layers. These convolution layers learn representative features of input images and construct segmentations based on the features. However, the features learned by standard convolution layers are not distinctive when the differences among different categories are subtle in terms of intensity, location, shape, and size. In this paper, we propose a novel CNN architecture, called Dense-Res-Inception Net (DRINet), which addresses this challenging problem. The proposed DRINet consists of three blocks, namely a convolutional block with dense connections, a deconvolutional block with residual inception modules, and an unpooling block. Our proposed architecture outperforms the U-Net in three different challenging applications, namely multi-class segmentation of cerebrospinal fluid on brain CT images, multi-organ segmentation on abdominal CT images, and multi-class brain tumor segmentation on MR images.

Limited capture range, and the requirement to provide high quality initialization for optimization-based 2-D/3-D image registration methods, can significantly degrade the performance of 3-D image reconstruction and motion compensation pipelines. Challenging clinical imaging scenarios, which contain significant subject motion, such as fetal in-utero imaging, complicate the 3-D image and volume reconstruction process. In this paper, we present a learning-based image registration method capable of predicting 3-D rigid transformations of arbitrarily oriented 2-D image slices, with respect to a learned canonical atlas co-ordinate system. Only image slice intensity information is used to perform registration and canonical alignment, no spatial transform initialization is required. To find image transformations, we utilize a convolutional neural network architecture to learn the regression function capable of mapping 2-D image slices to a 3-D canonical atlas space. We extensively evaluate the effectiveness of our approach quantitatively on simulated magnetic resonance imaging (MRI), fetal brain imagery with synthetic motion and further demonstrate qualitative results on real fetal MRI data where our method is integrated into a full reconstruction and motion compensation pipeline. Our learning based registration achieves an average spatial prediction error of 7 mm on simulated data and produces qualitatively improved reconstructions for heavily moving fetuses with gestational ages of approximately 20 weeks. Our model provides a general and computationally efficient solution to the 2-D/3-D registration initialization problem and is suitable for real-time scenarios.

Geometric distortion induced by the main B0 field disrupts the consistency of fetal echo planar imaging (EPI) data, on which diffusion and functional magnetic resonance imaging is based. In this paper, we present a novel data-driven method for simultaneous motion and distortion correction of fetal EPI. A motion-corrected and reconstructed T2 weighted single shot fast spin echo (ssFSE) volume is used as a model of undistorted fetal brain anatomy. Our algorithm interleaves two registration steps: estimation of fetal motion parameters by aligning EPI slices to the model; and deformable registration of EPI slices to slices simulated from the undistorted model to estimate the distortion field. The deformable registration is regularized by a physically inspired Laplacian constraint, to model distortion induced by a source-free background B0 field. Our experiments show that distortion correction significantly improves consistency of reconstructed EPI volumes with ssFSE volumes. In addition, the estimated distortion fields are consistent with fields calculated from acquired field maps, and the Laplacian constraint is essential for estimation of plausible distortion fields. The EPI volumes reconstructed from different scans of the same subject were more consistent when the proposed method was used in comparison with EPI volumes reconstructed from data distortion corrected using a separately acquired B0 field map.

Incorporation of prior knowledge about organ shape and location is key to improve performance of image analysis approaches. In particular, priors can be useful in cases where images are corrupted and contain artefacts due to limitations in image acquisition. The highly constrained nature of anatomical objects can be well captured with learning-based techniques. However, in most recent and promising techniques such as CNN-based segmentation it is not obvious how to incorporate such prior knowledge. State-of-the-art methods operate as pixel-wise classifiers where the training objectives do not incorporate the structure and inter-dependencies of the output. To overcome this limitation, we propose a generic training strategy that incorporates anatomical prior knowledge into CNNs through a new regularisation model, which is trained end-to-end. The new framework encourages models to follow the global anatomical properties of the underlying anatomy (e.g. shape, label structure) via learnt non-linear representations of the shape. We show that the proposed approach can be easily adapted to different analysis tasks (e.g. image enhancement, segmentation) and improve the prediction accuracy of the state-of-the-art models. The applicability of our approach is shown on multi-modal cardiac data sets and public benchmarks. In addition, we demonstrate how the learnt deep models of 3-D shapes can be interpreted and used as biomarkers for classification of cardiac pathologies.

Inspired by recent advances in deep learning, we propose a framework for reconstructing dynamic sequences of 2-D cardiac magnetic resonance (MR) images from undersampled data using a deep cascade of convolutional neural networks (CNNs) to accelerate the data acquisition process. In particular, we address the case where data are acquired using aggressive Cartesian undersampling. First, we show that when each 2-D image frame is reconstructed independently, the proposed method outperforms state-of-the-art 2-D compressed sensing approaches, such as dictionary learning-based MR image reconstruction, in terms of reconstruction error and reconstruction speed. Second, when reconstructing the frames of the sequences jointly, we demonstrate that CNNs can learn spatio-temporal correlations efficiently by combining convolution and data sharing approaches. We show that the proposed method consistently outperforms state-of-the-art methods and is capable of preserving anatomical structure more faithfully up to 11-fold undersampling. Moreover, reconstruction is very fast: each complete dynamic sequence can be reconstructed in less than 10 s and, for the 2-D case, each image frame can be reconstructed in 23 ms, enabling real-time applications.

Respiratory motion poses significant challenges in image-guided interventions. In emerging treatments such as MR-guided HIFU or MR-guided radiotherapy, it may cause significant misalignments between interventional road maps obtained pre-procedure and the anatomy during the treatment, and may affect intra-procedural imaging such as MR-thermometry. Patient specific respiratory motion models provide a solution to this problem. They establish a correspondence between the patient motion and simpler surrogate data which can be acquired easily during the treatment. Patient motion can then be estimated during the treatment by acquiring only the simpler surrogate data. In the majority of classical motion modelling approaches once the correspondence between the surrogate data and the patient motion is established it cannot be changed unless the model is recalibrated. However, breathing patterns are known to significantly change in the time frame of MR-guided interventions. Thus, the classical motion modelling approach may yield inaccurate motion estimations when the relation between the motion and the surrogate data changes over the duration of the treatment and frequent recalibration may not be feasible. We propose a novel methodology for motion modelling which has the ability to automatically adapt to new breathing patterns. This is achieved by choosing the surrogate data in such a way that it can be used to estimate the current motion in 3D as well as to update the motion model. In particular, in this work, we use 2D MR slices from different slice positions to build as well as to apply the motion model. We implemented such an autoadaptive motion model by extending our previous work on manifold alignment. We demonstrate a proof-of-principle of the proposed technique on cardiac gated data of the thorax and evaluate its adaptive behaviour on realistic synthetic data containing two breathing types generated from 6 volunteers, and real data from 4 volunteers. On synthetic data the autoadaptive motion model yielded 21.45% more accurate motion estimations compared to a non-adaptive motion model 10 min after a change in breathing pattern. On real data we demonstrated the method's ability to maintain motion estimation accuracy despite a drift in the respiratory baseline. Due to the cardiac gating of the imaging data, the method is currently limited to one update per heart beat and the calibration requires approximately 12 min of scanning. Furthermore, the method has a prediction latency of 800 ms. These limitations may be overcome in future work by altering the acquisition protocol.

We propose a dual pathway, 11-layers deep, three-dimensional Convolutional Neural Network for the challenging task of brain lesion segmentation. The devised architecture is the result of an in-depth analysis of the limitations of current networks proposed for similar applications. To overcome the computational burden of processing 3D medical scans, we have devised an efficient and effective dense training scheme which joins the processing of adjacent image patches into one pass through the network while automatically adapting to the inherent class imbalance present in the data. Further, we analyze the development of deeper, thus more discriminative 3D CNNs. In order to incorporate both local and larger contextual information, we employ a dual pathway architecture that processes the input images at multiple scales simultaneously. For post-processing of the network's soft segmentation, we use a 3D fully connected Conditional Random Field which effectively removes false positives. Our pipeline is extensively evaluated on three challenging tasks of lesion segmentation in multi-channel MRI patient data with traumatic brain injuries, brain tumours, and ischemic stroke. We improve on the state-of-the-art for all three applications, with top ranking performance on the public benchmarks BRATS 2015 and ISLES 2015. Our method is computationally efficient, which allows its adoption in a variety of research and clinical settings. The source code of our implementation is made publicly available.

We present a framework for combining a cardiac motion atlas with non-motion data. The atlas represents cardiac cycle motion across a number of subjects in a common space based on rich motion descriptors capturing 3D displacement, velocity, strain and strain rate. The non-motion data are derived from a variety of sources such as imaging, electrocardiogram (ECG) and clinical reports. Once in the atlas space, we apply a novel supervised learning approach based on random projections and ensemble learning to learn the relationship between the atlas data and some desired clinical output. We apply our framework to the problem of predicting response to Cardiac Resynchronisation Therapy (CRT). Using a cohort of 34 patients selected for CRT using conventional criteria, results show that the combination of motion and non-motion data enables CRT response to be predicted with 91.2% accuracy (100% sensitivity and 62.5% specificity), which compares favourably with the current state-of-the-art in CRT response prediction.

In this paper, we present a novel method for the correction of motion artifacts that are present in fetal magnetic resonance imaging (MRI) scans of the whole uterus. Contrary to current slice-to-volume registration (SVR) methods, requiring an inflexible anatomical enclosure of a single investigated organ, the proposed patch-to-volume reconstruction (PVR) approach is able to reconstruct a large field of view of non-rigidly deforming structures. It relaxes rigid motion assumptions by introducing a specific amount of redundant information that is exploited with parallelized patchwise optimization, super-resolution, and automatic outlier rejection. We further describe and provide an efficient parallel implementation of PVR allowing its execution within reasonable time on commercially available graphics processing units, enabling its use in the clinical practice. We evaluate PVR's computational overhead compared with standard methods and observe improved reconstruction accuracy in the presence of affine motion artifacts compared with conventional SVR in synthetic experiments. Furthermore, we have evaluated our method qualitatively and quantitatively on real fetal MRI data subject to maternal breathing and sudden fetal movements. We evaluate peak-signal-to-noise ratio, structural similarity index, and cross correlation with respect to the originally acquired data and provide a method for visual inspection of reconstruction uncertainty. We further evaluate the distance error for selected anatomical landmarks in the fetal head, as well as calculating the mean and maximum displacements resulting from automatic non-rigid registration to a motion-free ground truth image. These experiments demonstrate a successful application of PVR motion compensation to the whole fetal body, uterus, and placenta.

Identifying and interpreting fetal standard scan planes during 2-D ultrasound mid-pregnancy examinations are highly complex tasks, which require years of training. Apart from guiding the probe to the correct location, it can be equally difficult for a non-expert to identify relevant structures within the image. Automatic image processing can provide tools to help experienced as well as inexperienced operators with these tasks. In this paper, we propose a novel method based on convolutional neural networks, which can automatically detect 13 fetal standard views in freehand 2-D ultrasound data as well as provide a localization of the fetal structures via a bounding box. An important contribution is that the network learns to localize the target anatomy using weak supervision based on image-level labels only. The network architecture is designed to operate in real-time while providing optimal output for the localization task. We present results for real-time annotation, retrospective frame retrieval from saved videos, and localization on a very large and challenging dataset consisting of images and video recordings of full clinical anomaly screenings. We found that the proposed method achieved an average F1-score of 0.798 in a realistic classification experiment modeling real-time detection, and obtained a 90.09% accuracy for retrospective frame retrieval. Moreover, an accuracy of 77.8% was achieved on the localization task.

Accurate localization of anatomical landmarks is an important step in medical imaging, as it provides useful prior information for subsequent image analysis and acquisition methods. It is particularly useful for initialization of automatic image analysis tools (e.g. segmentation and registration) and detection of scan planes for automated image acquisition. Landmark localization has been commonly performed using learning based approaches, such as classifier and/or regressor models. However, trained models may not generalize well in heterogeneous datasets when the images contain large differences due to size, pose and shape variations of organs. To learn more data-adaptive and patient specific models, we propose a novel stratification based training model, and demonstrate its use in a decision forest. The proposed approach does not require any additional training information compared to the standard model training procedure and can be easily integrated into any decision tree framework. The proposed method is evaluated on 1080 3D high-resolution and 90 multi-stack 2D cardiac cine MR images. The experiments show that the proposed method achieves state-of-the-art landmark localization accuracy and outperforms standard regression and classification based approaches. Additionally, the proposed method is used in a multi-atlas segmentation to create a fully automatic segmentation pipeline, and the results show that it achieves state-of-the-art segmentation accuracy.

In this paper, we propose DeepCut, a method to obtain pixelwise object segmentations given an image dataset labelled weak annotations, in our case bounding boxes. It extends the approach of the well-known GrabCut [1] method to include machine learning by training a neural network classifier from bounding box annotations. We formulate the problem as an energy minimisation problem over a densely-connected conditional random field and iteratively update the training targets to obtain pixelwise object segmentations. Additionally, we propose variants of the DeepCut method and compare those to a naïve approach to CNN training under weak supervision. We test its applicability to solve brain and lung segmentation problems on a challenging fetal magnetic resonance dataset and obtain encouraging results in terms of accuracy.

When integrating computational tools, such as automatic segmentation, into clinical practice, it is of utmost importance to be able to assess the level of accuracy on new data and, in particular, to detect when an automatic method fails. However, this is difficult to achieve due to the absence of ground truth. Segmentation accuracy on clinical data might be different from what is found through cross validation, because validation data are often used during incremental method development, which can lead to overfitting and unrealistic performance expectations. Before deployment, performance is quantified using different metrics, for which the predicted segmentation is compared with a reference segmentation, often obtained manually by an expert. But little is known about the real performance after deployment when a reference is unavailable. In this paper, we introduce the concept of reverse classification accuracy (RCA) as a framework for predicting the performance of a segmentation method on new data. In RCA, we take the predicted segmentation from a new image to train a reverse classifier, which is evaluated on a set of reference images with available ground truth. The hypothesis is that if the predicted segmentation is of good quality, then the reverse classifier will perform well on at least some of the reference images. We validate our approach on multi-organ segmentation with different classifiers and segmentation methods. Our results indicate that it is indeed possible to predict the quality of individual segmentations, in the absence of ground truth. Thus, RCA is ideal for integration into automatic processing pipelines in clinical routine and as a part of large-scale image analysis studies.

Recently, several models based on deep neural networks have achieved great success in terms of both reconstruction accuracy and computational performance for single image super-resolution. In these methods, the low resolution (LR) input image is upscaled to the high resolution (HR) space using a single filter, commonly bicubic interpolation, before reconstruction. This means that the super-resolution (SR) operation is performed in HR space. We demonstrate that this is sub-optimal and adds computational complexity. In this paper, we present the first convolutional neural network (CNN) capable of real-time SR of 1080p videos on a single K2 GPU. To achieve this, we propose a novel CNN architecture where the feature maps are extracted in the LR space. In addition, we introduce an efficient sub-pixel convolution layer which learns an array of upscaling filters to upscale the final LR feature maps into the HR output. By doing so, we effectively replace the handcrafted bicubic filter in the SR pipeline with more complex upscaling filters specifically trained for each feature map, whilst also reducing the computational complexity of the overall SR operation. We evaluate the proposed approach using images and videos from publicly available datasets and show that it performs significantly better (+0.15dB on Images and +0.39dB on Videos) and is an order of magnitude faster than previous CNN-based methods.

Face-space has become established as an effective model for representing the dimensions of variation that occur in collections of human faces. For example, a change of expression from neutral to smiling can be represented by one axis in a face space. Principal components can be used to determine the axes of a face-space, however, standard principal components are based entirely on the data set from which they are computed, and do not express any domain specific information about the application of interest. In this paper, we propose a face-space analysis that combines the variance criterion used in principal components with some prior knowledge about the task-driven experiment. The priors are based on measuring eye movements of participants to frontal 2D faces during separate gender and facial expression categorization tasks. Our findings show that saccades to faces are task-driven, especially from 500 to 1000 milliseconds, and automatic recognition performance does not improve with additional exposure time.

3D cardiac MR imaging enables accurate analysis of cardiac morphology and physiology. However, due to the requirements for long acquisition and breath-hold, the clinical routine is still dominated by multi-slice 2D imaging, which hamper the visualization of anatomy and quantitative measurements as relatively thick slices are acquired. As a solution, we propose a novel image super-resolution (SR) approach that is based on a residual convolutional neural network (CNN) model. It reconstructs high resolution 3D volumes from 2D image stacks for more accurate image analysis. The proposed model allows the use of multiple input data acquired from different viewing planes for improved performance. Experimental results on 1233 cardiac short and long-axis MR image stacks show that the CNN model outperforms state-of-the-art SR methods in terms of image quality while being computationally efficient. Also, we show that image segmentation and motion tracking benefits more from SR-CNN when it is used as an initial upscaling method than conventional interpolation methods for the subsequent analysis.

Image registration under challenging realistic conditions is a very important area of research. In this paper, we focus on algorithms that seek to densely align two volumetric images according to a global similarity measure. Despite intensive research in this area, there is still a need for similarity measures that are robust to outliers common to many different types of images. For example, medical image data is often corrupted by intensity inhomogeneities and may contain outliers in the form of pathologies. In this paper we propose a global similarity measure that is robust to both intensity inhomogeneities and outliers without requiring prior knowledge of the type of outliers. We combine the normalised gradients of images with the cosine function and show that it is theoretically robust against a very general class of outliers. Experimentally, we verify the robustness of our measures within two distinct algorithms. Firstly, we embed our similarity measures within a proof-of-concept extension of the Lucas–Kanade algorithm for volumetric data. Finally, we embed our measures within a popular non-rigid alignment framework based on free-form deformations and show it to be robust against both simulated tumours and intensity inhomogeneities.

Real-time 3D Echocardiography (RT3DE) has been proven to be an accurate tool for left ventricular (LV) volume assessment. However, identification of the LV endocardium remains a challenging task, mainly because of the low tissue/blood contrast of the images combined with typical artifacts. Several semi and fully automatic algorithms have been proposed for segmenting the endocardium in RT3DE data in order to extract relevant clinical indices, but a systematic and fair comparison between such methods has so far been impossible due to the lack of a publicly available common database. Here, we introduce a standardized evaluation framework to reliably evaluate and compare the performance of the algorithms developed to segment the LV border in RT3DE. A database consisting of 45 multivendor cardiac ultrasound recordings acquired at different centers with corresponding reference measurements from three experts are made available. The algorithms from nine research groups were quantitatively evaluated and compared using the proposed online platform. The results showed that the best methods produce promising results with respect to the experts' measurements for the extraction of clinical indices, and that they offer good segmentation precision in terms of mean distance error in the context of the experts' variability range. The platform remains open for new submissions.

We propose a framework for the robust and fully-automatic segmentation of magnetic resonance (MR) brain images called "Multi-Atlas Label Propagation with Expectation-Maximisation based refinement" (MALP-EM). The presented approach is based on a robust registration approach (MAPER), highly performant label fusion (joint label fusion) and intensity-based label refinement using EM. We further adapt this framework to be applicable for the segmentation of brain images with gross changes in anatomy. We propose to account for consistent registration errors by relaxing anatomical priors obtained by multi-atlas propagation and a weighting scheme to locally combine anatomical atlas priors and intensity-refined posterior probabilities. The method is evaluated on a benchmark dataset used in a recent MICCAI segmentation challenge. In this context we show that MALP-EM is competitive for the segmentation of MR brain scans of healthy adults when compared to state-of-the-art automatic labelling techniques. To demonstrate the versatility of the proposed approach, we employed MALP-EM to segment 125 MR brain images into 134 regions from subjects who had sustained traumatic brain injury (TBI). We employ a protocol to assess segmentation quality if no manual reference labels are available. Based on this protocol, three independent, blinded raters confirmed on 13 MR brain scans with pathology that MALP-EM is superior to established label fusion techniques. We visually confirm the robustness of our segmentation approach on the full cohort and investigate the potential of derived symmetry-based imaging biomarkers that correlate with and predict clinically relevant variables in TBI such as the Marshall Classification (MC) or Glasgow Outcome Score (GOS). Specifically, we show that we are able to stratify TBI patients with favourable outcomes from non-favourable outcomes with 64.7% accuracy using acute-phase MR images and 66.8% accuracy using follow-up MR images. Furthermore, we are able to differentiate subjects with the presence of a mass lesion or midline shift from those with diffuse brain injury with 76.0% accuracy. The thalamus, putamen, pallidum and hippocampus are particularly affected. Their involvement predicts TBI disease progression.

An automated segmentation method is presented for multi-organ segmentation in abdominal CT images. Dictionary learning and sparse coding techniques are used in the proposed method to generate target specific priors for segmentation. The method simultaneously learns dictionaries which have reconstructive power and classifiers which have discriminative ability from a set of selected atlases. Based on the learnt dictionaries and classifiers, probabilistic atlases are then generated to provide priors for the segmentation of unseen target images. The final segmentation is obtained by applying a post-processing step based on a graph-cuts method. In addition, this paper proposes a voxel-wise local atlas selection strategy to deal with high inter-subject variation in abdominal CT images. The segmentation performance of the proposed method with different atlas selection strategies are also compared. Our proposed method has been evaluated on a database of 150 abdominal CT images and achieves a promising segmentation performance with Dice overlap values of 94.9%, 93.6%, 71.1%, and 92.5% for liver, kidneys, pancreas, and spleen, respectively.

Clinical annotations, such as voxel-wise binary or probabilistic tissue segmentations, structural parcellations, pathological regions-of-interest and anatomical landmarks are key to many clinical studies. However, due to the time consuming nature of manually generating these annotations, they tend to be scarce and limited to small subsets of data. This work explores a novel framework to propagate voxel-wise annotations between morphologically dissimilar images by diffusing and mapping the available examples through intermediate steps. A spatially-variant graph structure connecting morphologically similar subjects is introduced over a database of images, enabling the gradual diffusion of information to all the subjects, even in the presence of large-scale morphological variability. We illustrate the utility of the proposed framework on two example applications: brain parcellation using categorical labels and tissue segmentation using probabilistic features. The application of the proposed method to categorical label fusion showed highly statistically significant improvements when compared to state-of-the-art methodologies. Significant improvements were also observed when applying the proposed framework to probabilistic tissue segmentation of both synthetic and real data, mainly in the presence of large morphological variability.

We demonstrate a new method to recover 4D blood flow over the entire ventricle from partial blood velocity measurements using multiple 3D+t colour Doppler images and ventricular wall motion estimated using 3D+t BMode images. We apply our approach to realistic simulated data to ascertain the ability of the method to deal with incomplete data, as typically happens in clinical practice. Experiments using synthetic data show that the use of wall motion improves velocity reconstruction, shows more accurate flow patterns and improves mean accuracy particularly when coverage of the ventricle is poor. The method was applied to patient data from 6 congenital cases, producing results consistent with the simulations. The use of wall motion produced more plausible flow patterns and reduced the reconstruction error in all patients.

Capturing an enclosing volume of moving subjects and organs using fast individual image slice acquisition has shown promise in dealing with motion artefacts. Motion between slice acquisitions results in spatial inconsistencies that can be resolved by slice-to-volume reconstruction (SVR) methods to provide high quality 3D image data. Existing algorithms are, however, typically very slow, specialised to specific applications and rely on approximations, which impedes their potential clinical use. In this paper, we present a fast multi-GPU accelerated framework for slice-to-volume reconstruction. It is based on optimised 2D/3D registration, super-resolution with automatic outlier rejection and an additional (optional) intensity bias correction. We introduce a novel and fully automatic procedure for selecting the image stack with least motion to serve as an initial registration target. We evaluate the proposed method using artificial motion corrupted phantom data as well as clinical data, including tracked freehand ultrasound of the liver and fetal Magnetic Resonance Imaging. We achieve speed-up factors greater than 30 compared to a single CPU system and greater than 10 compared to currently available state-of-the-art multi-core CPU methods. We ensure high reconstruction accuracy by exact computation of the point-spread function for every input data point, which has not previously been possible due to computational limitations. Our framework and its implementation is scalable for available computational infrastructures and tests show a speed-up factor of 1.70 for each additional GPU. This paves the way for the online application of image based reconstruction methods during clinical examinations. The source code for the proposed approach is publicly available.

Echocardiography is a potential alternative to X-ray fluoroscopy in cardiac catheterization given its richness in soft tissue information and its lack of ionizing radiation. However, its small field of view and acoustic artifacts make direct automatic segmentation of the catheters very challenging. In this study, a fast catheter segmentation framework for echocardiographic imaging guided by the segmentation of corresponding X-ray fluoroscopic imaging is proposed. The complete framework consists of: 1) catheter initialization in the first X-ray frame; 2) catheter tracking in the rest of the X-ray sequence; 3) fast registration of corresponding X-ray and ultrasound frames; and 4) catheter segmentation in ultrasound images guided by the results of both X-ray tracking and fast registration. The main contributions include: 1) a Kalman filter-based growing strategy with more clinical data evalution; 2) a SURF detector applied in a constrained search space for catheter segmentation in ultrasound images; 3) a two layer hierarchical graph model to integrate and smooth catheter fragments into a complete catheter; and 4) the integration of these components into a system for clinical applications. This framework is evaluated on five sequences of porcine data and four sequences of patient data comprising more than 3000 X-ray frames and more than 1000 ultrasound frames. The results show that our algorithm is able to track the catheter in ultrasound images at 1.3 s per frame, with an error of less than 2 mm. However, although this may satisfy the accuracy for visualization purposes and is also fast, the algorithm still needs to be further accelerated for real-time clinical applications.

The quantification of similarity between image segmentations is a complex yet important task. The ideal similarity measure should be unbiased to segmentations of different volume and complexity, and be able to quantify and visualise segmentation bias. Similarity measures based on overlap, e.g. Dice score, or surface distances, e.g. Hausdorff distance, clearly do not satisfy all of these properties. To address this problem, we introduce Patch-based Evaluation of Image Segmentation (PEIS), a general method to assess segmentation quality. Our method is based on finding patch correspondences and the associated patch displacements, which allow the estimation of segmentation bias. We quantify both the agreement of the segmentation boundary and the conservation of the segmentation shape. We further assess the segmentation complexity within patches to weight the contribution of local segmentation similarity to the global score. We evaluate PEIS on both synthetic data and two medical imaging datasets. On synthetic segmentations of different shapes, we provide evidence that PEIS, in comparison to the Dice score, produces more comparable scores, has increased sensitivity and estimates segmentation bias accurately. On cardiac magnetic resonance (MR) images, we demonstrate that PEIS can evaluate the performance of a segmentation method independent of the size or complexity of the segmentation under consideration. On brain MR images, we compare five different automatic hippocampus segmentation techniques using PEIS. Finally, we visualise the segmentation bias on a selection of the cases.

We propose a fully automatic learning-based multi-atlas approach to segment the prostate using multi-channel (T1 and T2) MR images. After affine transformation to the template space, multi-scale features are extracted and separate random forest classifiers are learnt for the prostate region from the most similar T1 and T2 atlases. The probabilities from these two classifiers (T1 and T2) are then fused to obtain a robust probabilistic atlas. Finally, using the probabilistic representation for each voxel, the multi-image graph cuts algorithm is applied on these multi-channel images simultaneously to get the final segmentation. The novelty of the proposed method lies in the use of multi-channel MR images, a decision forest learnt from only the most similar MR images, and the fusion of global and local template-based classifiers for prostate segmentation. We apply this method to a set of 107 prostate images, with 77 randomly selected images used for training and the remaining 30 images for testing. The results are compared to the radiologist's labeled ground truth using cross-validation. The best result is obtained via hybrid approach in which the global classifier trained on T1 images and local template-based classifiers trained on T2 images are fused to obtain the final probability for each voxel. Our results indicate that the proposed method is robust, capable of producing accurate segmentation automatically and most importantly, not patient-specific.

We present a novel method of hierarchical manifold learning which aims to automatically discover regional properties of image datasets. While traditional manifold learning methods have become widely used for dimensionality reduction in medical imaging, they suffer from only being able to consider whole images as single data points. We extend conventional techniques by additionally examining local variations, in order to produce spatially-varying manifold embeddings that characterize a given dataset. This involves constructing manifolds in a hierarchy of image patches of increasing granularity, while ensuring consistency between hierarchy levels. We demonstrate the utility of our method in two very different settings: 1) to learn the regional correlations in motion within a sequence of time-resolved MR images of the thoracic cavity; 2) to find discriminative regions of 3-D brain MR images associated with neurodegenerative disease.

The reconstruction of dynamic magnetic resonance data from an undersampled k-space has been shown to have a huge potential in accelerating the acquisition process of this imaging modality. With the introduction of compressed sensing (CS) theory, solutions for undersampled data have arisen which reconstruct images consistent with the acquired samples and compliant with a sparsity model in some transform domain. Fixed basis transforms have been extensively used as sparsifying transforms in the past, but recent developments in dictionary learning (DL) have been shown to outperform them by training an overcomplete basis that is optimal for a particular dataset. We present here an iterative algorithm that enables the application of DL for the reconstruction of cardiac cine data with Cartesian undersampling. This is achieved with local processing of spatio-temporal 3D patches and by independent treatment of the real and imaginary parts of the dataset. The enforcement of temporal gradients is also proposed as an additional constraint that can greatly accelerate the convergence rate and improve the reconstruction for high acceleration rates. The method is compared to and shown to systematically outperform k- t FOCUSS, a successful CS method that uses a fixed basis transform.

Magnetic resonance (MR) imaging is increasingly being used to assess brain growth and development in infants. Such studies are often based on quantitative analysis of anatomical segmentations of brain MR images. However, the large changes in brain shape and appearance associated with development, the lower signal to noise ratio and partial volume effects in the neonatal brain present challenges for automatic segmentation of neonatal MR imaging data. In this study, we propose a framework for accurate intensity-based segmentation of the developing neonatal brain, from the early preterm period to term-equivalent age, into 50 brain regions. We present a novel segmentation algorithm that models the intensities across the whole brain by introducing a structural hierarchy and anatomical constraints. The proposed method is compared to standard atlas-based techniques and improves label overlaps with respect to manual reference segmentations. We demonstrate that the proposed technique achieves highly accurate results and is very robust across a wide range of gestational ages, from 24 weeks gestational age to term-equivalent age.

An unresolved issue in patients with diastolic dysfunction is that the estimation of myocardial stiffness cannot be decoupled from diastolic residual active tension (AT) because of the impaired ventricular relaxation during diastole. To address this problem, this paper presents a method for estimating diastolic mechanical parameters of the left ventricle (LV) from cine and tagged MRI measurements and LV cavity pressure recordings, separating the passive myocardial constitutive properties and diastolic residual AT. Dynamic C1-continuous meshes are automatically built from the anatomy and deformation captured from dynamic MRI sequences. Diastolic deformation is simulated using a mechanical model that combines passive and active material properties. The problem of non-uniqueness of constitutive parameter estimation using the well known Guccione law is characterized by reformulation of this law. Using this reformulated form, and by constraining the constitutive parameters to be constant across time points during diastole, we separate the effects of passive constitutive properties and the residual AT during diastolic relaxation. Finally, the method is applied to two clinical cases and one control, demonstrating that increased residual AT during diastole provides a potential novel index for delineating healthy and pathological cases.

The evaluation of ventricular function is important for the diagnosis of cardiovascular diseases. It typically involves measurement of the left ventricular (LV) mass and LV cavity volume. Manual delineation of the myocardial contours is time-consuming and dependent on the subjective experience of the expert observer. In this paper, a multi-atlas method is proposed for cardiac magnetic resonance (MR) image segmentation. The proposed method is novel in two aspects. First, it formulates a patch-based label fusion model in a Bayesian framework. Second, it improves image registration accuracy by utilizing label information, which leads to improvement of segmentation accuracy. The proposed method was evaluated on a cardiac MR image set of 28 subjects. The average Dice overlap metric of our segmentation is 0.92 for the LV cavity, 0.89 for the right ventricular cavity and 0.82 for the myocardium. The results show that the proposed method is able to provide accurate information for clinical diagnosis.

Functional connections between brain regions are supported by structural connectivity. Both functional and structural connectivity are estimated from in vivo magnetic resonance imaging and offer complementary information on brain organization and function. However, imaging only provides noisy measures, and we lack a good neuroscientific understanding of the links between structure and function. Therefore, inter-subject joint modeling of structural and functional connectivity, the key to multimodal biomarkers, is an open challenge. We present a probabilistic framework to learn across subjects a mapping from structural to functional brain connectivity. Expanding on our previous work [1], our approach is based on a predictive framework with multiple sparse linear regression. We rely on the randomized LASSO to identify relevant anatomo-functional links with some confidence interval. In addition, we describe resting-state functional magnetic resonance imaging in the setting of Gaussian graphical models, on the one hand imposing conditional independences from structural connectivity and on the other hand parameterizing the problem in terms of multivariate autoregressive models. We introduce an intrinsic measure of prediction error for functional connectivity that is independent of the parameterization chosen and provides the means for robust model selection. We demonstrate our methodology with regions within the default mode and the salience network as well as, atlas-based cortical parcellation.

A robust automated segmentation of abdominal organs can be crucial for computer aided diagnosis and laparoscopic surgery assistance. Many existing methods are specialized to the segmentation of individual organs and struggle to deal with the variability of the shape and position of abdominal organs. We present a general, fully-automated method for multi-organ segmentation of abdominal computed tomography (CT) scans. The method is based on a hierarchical atlas registration and weighting scheme that generates target specific priors from an atlas database by combining aspects from multi-atlas registration and patch-based segmentation, two widely used methods in brain segmentation. The final segmentation is obtained by applying an automatically learned intensity model in a graph-cuts optimization step, incorporating high-level spatial knowledge. The proposed approach allows to deal with high inter-subject variation while being flexible enough to be applied to different organs. We have evaluated the segmentation on a database of 150 manually segmented CT images. The achieved results compare well to state-of-the-art methods, that are usually tailored to more specific questions, with Dice overlap values of 94%, 93%, 70%, and 92% for liver, kidneys, pancreas, and spleen, respectively.

In this paper, we present a novel technique based on nonrigid image registration for myocardial motion estimation using both untagged and 3-D tagged MR images. The novel aspect of our technique is its simultaneous usage of complementary information from both untagged and 3-D tagged MR images. To estimate the motion within the myocardium, we register a sequence of tagged and untagged MR images during the cardiac cycle to a set of reference tagged and untagged MR images at end-diastole. The similarity measure is spatially weighted to maximize the utility of information from both images. In addition, the proposed approach integrates a valve plane tracker and adaptive incompressibility into the framework. We have evaluated the proposed approach on 12 subjects. Our results show a clear improvement in terms of accuracy compared to approaches that use either 3-D tagged or untagged MR image information alone. The relative error compared to manually tracked landmarks is less than 15% throughout the cardiac cycle. Finally, we demonstrate the automatic analysis of cardiac function from the myocardial deformation fields.

In this paper we propose a method that exploits 3D motion-based features between frames of 3D facial geometry sequences for dynamic facial expression recognition. An expressive sequence is modeled to contain an onset followed by an apex and an offset. Feature selection methods are applied in order to extract features for each of the onset and offset segments of the expression. These features are then used to train a Hidden Markov Model in order to model the full temporal dynamics of the expression. The proposed fully automatic system was tested in a subset of the BU-4DFE database for the recognition of happiness, anger and surprise. Comparisons with a similar system based on the motion extracted from facial intensity images was also performed. The attained results suggest that the use of the 3D information does indeed improve the recognition accuracy when compared to the 2D data.

Large medical image datasets form a rich source of anatomical descriptions for research into pathology and clinical biomarkers. Many features may be extracted from data such as MR images to provide, through manifold learning methods, new representations of the population's anatomy. However, the ability of any individual feature to fully capture all aspects morphology is limited. We propose a framework for deriving a representation from multiple features or measures which can be chosen to suit the application and are processed using separate manifold-learning steps. The results are then combined to give a single set of embedding coordinates for the data. We illustrate the framework in a population study of neonatal brain MR images and show how consistent representations, correlating well with clinical data, are given by measures of shape and of appearance. These particular measures were chosen as the developing neonatal brain undergoes rapid changes in shape and MR appearance and were derived from extracted cortical surfaces, nonrigid deformations, and image similarities. Combined single embeddings show improved correlations demonstrating their benefit for further studies such as identifying patterns in the trajectories of brain development. The results also suggest a lasting effect of age at birth on brain morphology, coinciding with previous clinical studies.

In the framework of large deformation diffeomorphic metric mapping (LDDMM), we present a practical methodology to integrate prior knowledge about the registered shapes in the regularizing metric. Our goal is to perform rich anatomical shape comparisons from volumetric images with the mathematical properties offered by the LDDMM framework. We first present the notion of characteristic scale at which image features are deformed. We then propose a methodology to compare anatomical shape variations in a multi-scale fashion, i.e., at several characteristic scales simultaneously. In this context, we propose a strategy to quantitatively measure the feature differences observed at each characteristic scale separately. After describing our methodology, we illustrate the performance of the method on phantom data. We then compare the ability of our method to segregate a group of subjects having Alzheimer's disease and a group of controls with a classical coarse to fine approach, on standard 3D MR longitudinal brain images. We finally apply the approach to quantify the anatomical development of the human brain from 3D MR longitudinal images of pre-term babies. Results show that our method registers accurately volumetric images containing feature differences at several scales simultaneously with smooth deformations.

Abstract In this paper we present a predictive performance model for a key biomedical imaging application found as part of the U.K. e‐Science Information eXtraction from Images (IXI) project. This code represents a significant challenge for our existing performance prediction tools as it has internal structures that exhibit highly variable runtimes depending on qualities in the input data provided. Since the runtime can vary by more than an order of magnitude, it has been difficult to apply meaningful quality of service criteria to workflows that use this code. The model developed here is used in the context of an interactive scheduling system which provides rapid feedback to the users, allowing them to tailor their workloads to available resources or to allocate extra resources to scheduled workloads. Copyright © 2007 John Wiley & Sons, Ltd.

Motion degrades magnetic resonance (MR) images and prevents acquisition of self-consistent and high-quality volume images. A novel methodology, Snapshot magnetic resonance imaging (MRI) with Volume Reconstruction (SVR) has been developed for imaging moving subjects at high resolution and high signal-to-noise ratio (SNR). The method combines registered 2-D slices from sequential dynamic single-shot scans. The SVR approach requires that the anatomy in question is not changing shape or size and is moving at a rate that allows snapshot images to be acquired. After imaging the target volume repeatedly to guarantee sufficient sampling every where, a robust slice-to-volume registration method has been implemented that achieves alignment of each slice within 0.3 mm in the examples tested. Multilevel scattered interpolation has been used to obtain high-fidelity reconstruction with root-mean-square (rms) error that is less than the noise level in the images. The SVR method has been performed successfully for brain studies on subjects that cannot stay still, and in some cases were moving substantially during scanning. For example, awake neonates, deliberately moved adults and, especially, on fetuses, for which no conventional high-resolution 3-D method is currently available. Fine structure of the in-utero fetal brain is clearly revealed for the first time and substantial SNR improvement is realized by having many individually acquired slices contribute to each voxel in the reconstructed image.

The thirteen articles in this special issue are devoted to mathematical analysis of biomedical imaging processes and systems.

Recent innovations in drug therapies have made it highly desirable to obtain sensitive biomarkers of disease progression that can be used to quantify the performance of candidate disease modifying drugs. In order to measure potential image-based biomarkers of disease progression in an experimental model of rheumatoid arthritis (RA), we present two different methods to automatically quantify changes in a bone in in-vivo serial magnetic resonance (MR) images from the model. Both methods are based on rigid and nonrigid image registration to perform the analysis. The first method uses segmentation propagation to delineate a bone from the serial MR images giving a global measure of temporal changes in bone volume. The second method uses rigid body registration to determine intensity change within a bone, and then maps these into a reference coordinate system using nonrigid registration. This gives a local measure of temporal changes in bone lesion volume. We detected significant temporal changes in local bone lesion volume in five out of eight identified candidate bone lesion regions, and significant difference in local bone lesion volume between male and female subjects in three out of eight candidate bone lesion regions. But the global bone volume was found to be fluctuating over time. Finally, we compare our findings with histology of the subjects and the manual segmentation of bone lesions.

Generation of digitally reconstructed radiographs (DRRs) is computationally expensive and is typically the rate-limiting step in the execution time of intensity-based two-dimensional to three-dimensional (2D-3D) registration algorithms. We address this computational issue by extending the technique of light field rendering from the computer graphics community. The extension of light fields, which we call attenuation fields (AFs), allows most of the DRR computation to be performed in a preprocessing step; after this precomputation step, DRRs can be generated substantially faster than with conventional ray casting. We derive expressions for the physical sizes of the two planes of an AF necessary to generate DRRs for a given X-ray camera geometry and all possible object motion within a specified range. Because an AF is a ray-based data structure, it is substantially more memory efficient than a huge table of precomputed DRRs because it eliminates the redundancy of replicated rays. Nonetheless, an AF can require substantial memory, which we address by compressing it using vector quantization. We compare DRRs generated using AFs (AF-DRRs) to those generated using ray casting (RC-DRRs) for a typical C-arm geometry and computed tomography images of several anatomic regions. They are quantitatively very similar: the median peak signal-to-noise ratio of AF-DRRs versus RC-DRRs is greater than 43 dB in all cases. We perform intensity-based 2D-3D registration using AF-DRRs and RC-DRRs and evaluate registration accuracy using gold-standard clinical spine image data from four patients. The registration accuracy and robustness of the two methods is virtually identical whereas the execution speed using AF-DRRs is an order of magnitude faster.

Tagged magnetic resonance imaging (MRI) is unique in its ability to noninvasively image the motion and deformation of the heart in vivo, but one of the fundamental reasons limiting its use in the clinical environment is the absence of automated tools to derive clinically useful information from tagged MR images. In this paper, we present a novel and fully automated technique based on nonrigid image registration using multilevel free-form deformations (MFFDs) for the analysis of myocardial motion using tagged MRI. The novel aspect of our technique is its integrated nature for tag localization and deformation field reconstruction using image registration and voxel based similarity measures. To extract the motion field within the myocardium during systole we register a sequence of images taken during systole to a set of reference images taken at end-diastole, maximizing the normalized mutual information between the images. We use both short-axis and long-axis images of the heart to estimate the full four-dimensional motion field within the myocardium. We also present validation results from data acquired from twelve volunteers.

A method is presented to interpolate between neighboring slices in a grey-scale tomographic data set. Spatial correspondence between adjacent slices is established using a nonrigid registration algorithm based on B-splines which optimizes the normalized mutual information similarity measure. Linear interpolation of the image intensities is then carried out along the directions calculated by the registration algorithm. The registration-based method is compared to both standard linear interpolation and shape-based interpolation in 20 tomographic data sets. Results show that the proposed method statistically significantly outperforms both linear and shape-based interpolation.

In this paper, we present a technique that can be used to transform the motion or deformation fields defined in the coordinate system of one subject into the coordinate system of another subject. Such a transformation accounts for the differences in the coordinate systems of the two subjects due to misalignment and size/shape variation, enabling the motion or deformation of each of the subjects to be directly quantitatively and qualitatively compared. The field transformation is performed by using a nonrigid registration algorithm to determine the intersubject coordinate system mapping from the first subject to the second subject. This fixes the relationship between the coordinate systems of the two subjects, and allows us to recover the deformation/motion vectors of the second subject for each corresponding point in the first subject. Since these vectors are still aligned with the coordinate system of the second subject, the inverse of the intersubject coordinate mapping is required to transform these vectors into the coordinate system of the first subject, and we approximate this inverse using a numerical line integral method. The accuracy of our numerical inversion technique is demonstrated using a synthetic example, after which we present applications of our method to sequences of cardiac and brain images.

This paper describes techniques for producing FPGA-based designs that support free-form deformation in medical image processing. The free-form deformation method is based on a B-spline algorithm for modelling three-dimensional deformable objects. Our design includes four optimisations. First, we transform a nested loop to eliminate conditional statements. Second, we adopt a customised number representation format in our implementation. Third, we store the values of a third-order B-spline model in lookup tables. Fourth, we pipeline the design to increase its throughput, and we also deploy multiple pipelines such that each covers a different subimage. Our design description, captured in the Handel-C language, is parameterisable at compile time to support a range of image resolutions and computational precisions. An implementation on a Xilinx XC2V6000 device at 67 MHz can run 3.2 times faster than an Intel Xeon-based PC at 2666 MHz.

We describe a registration and tracking technique to integrate cardiac X-ray images and cardiac magnetic resonance (MR) images acquired from a combined X-ray and MR interventional suite (XMR). Optical tracking is used to determine the transformation matrices relating MR image coordinates and X-ray image coordinates. Calibration of X-ray projection geometry and tracking of the X-ray C-arm and table enable three-dimensional (3-D) reconstruction of vessel centerlines and catheters from bi-plane X-ray views. We can, therefore, combine single X-ray projection images with registered projection MR images from a volume acquisition, and we can also display 3-D reconstructions of catheters within a 3-D or multi-slice MR volume. Registration errors were assessed using phantom experiments. Errors in the combined projection images (two-dimensional target registration error--TRE) were found to be 2.4 to 4.2 mm, and the errors in the integrated volume representation (3-D TRE) were found to be 4.6 to 5.1 mm. These errors are clinically acceptable for alignment of images of the great vessels and the chambers of the heart. Results are shown for two patients. The first involves overlay of a catheter used for invasive pressure measurements on an MR volume that provides anatomical context. The second involves overlay of invasive electrode catheters (including a basket catheter) on a tagged MR volume in order to relate electrophysiology to myocardial motion in a patient with an arrhythmia. Visual assessment of these results suggests the errors were of a similar magnitude to those obtained in the phantom measurements.

In this paper, we show how the concept of statistical deformation models (SDMs) can be used for the construction of average models of the anatomy and their variability. SDMs are built by performing a statistical analysis of the deformations required to map anatomical features in one subject into the corresponding features in another subject. The concept of SDMs is similar to statistical shape models (SSMs) which capture statistical information about shapes across a population, but offers several advantages over SSMs. First, SDMs can be constructed directly from images such as three-dimensional (3-D) magnetic resonance (MR) or computer tomography volumes without the need for segmentation which is usually a prerequisite for the construction of SSMs. Instead, a nonrigid registration algorithm based on free-form deformations and normalized mutual information is used to compute the deformations required to establish dense correspondences between the reference subject and the subjects in the population class under investigation. Second, SDMs allow the construction of an atlas of the average anatomy as well as its variability across a population of subjects. Finally, SDMs take the 3-D nature of the underlying anatomy into account by analysing dense 3-D deformation fields rather than only information about the surface shape of anatomical structures. We show results for the construction of anatomical models of the brain from the MR images of 25 different subjects. The correspondences obtained by the nonrigid registration are evaluated using anatomical landmark locations and show an average error of 1.40 mm at these anatomical landmark positions. We also demonstrate that SDMs can be constructed so as to minimize the bias toward the chosen reference subject.

This paper describes techniques for producing FPGA-based designs that support free-form deformation in medical image processing. The free-form deformation method is based on a B-spline algorithm for modelling three-dimensional deformable objects. Our design includes four optimisations. First, we store the values of a third-order B-spline model in lookup tables. Second, we adopt a customised number representation format in our implementation. Third, we transform a nested loop so that conditionals are moved outside the loop. Fourth, we pipeline the design to increase its throughput, and we also deploy multiple pipelines such that each covers a different image. Our design description, captured in the Handel-C language, is parameterisable at compile time to support a range of image resolutions and computational precisions. An implementation on a Xilinx XC2V6000 device would be capable of processing images of resolution up to 256 by 256 pixels in real time.

A novel method is introduced for the generation of landmarks for three-dimensional (3-D) shapes and the construction of the corresponding 3-D statistical shape models. Automatic landmarking of a set of manual segmentations from a class of shapes is achieved by 1) construction of an atlas of the class, 2) automatic extraction of the landmarks from the atlas, and 3) subsequent propagation of these landmarks to each example shape via a volumetric nonrigid registration technique using multiresolution B-spline deformations. This approach presents some advantages over previously published methods: it can treat multiple-part structures and requires less restrictive assumptions on the structure's topology. In this paper, we address the problem of building a 3-D statistical shape model of the left and right ventricle of the heart from 3-D magnetic resonance images. The average accuracy in landmark propagation is shown to be below 2.2 mm. This application demonstrates the robustness and accuracy of the method in the presence of large shape variability and multiple objects.

The authors propose a novel method to register two or more optical images to a 3D surface model. The potential applications of such a registration method could be in medicine for example, in image guided interventions, surveillance and identification, industrial inspection, or telemanipulation in remote or hostile environments. Registration is performed by optimizing a similarity measure with respect to the transformation parameters. We propose a novel similarity measure based on "photo-consistency." For each surface point, the similarity measure computes how consistent the corresponding optical image information in each view is with a lighting model. The relative pose of the optical images must be known. We validate the system using data from an optical-based surface reconstruction system and surfaces derived from magnetic resonance (MR) images of the human face. We test the accuracy and robustness of the system with respect to the number of video images, video image noise, errors in surface location and area, and complexity of the matched surfaces. We demonstrate the algorithm working on 10 further optical-based reconstructions of the human head and skin surfaces derived from MR images of the heads of five volunteers. Matching four optical images to a surface model produced a 3D error of between 1.45 and 1.59 mm, at a success rate of 100 percent, where the initial misregistration was up to 16 mm or degrees from the registration position.

In this paper we present a new approach for the nonrigid registration of contrast-enhanced breast MRI. A hierarchical transformation model of the motion of the breast has been developed. The global motion of the breast is modeled by an affine transformation while the local breast motion is described by a free-form deformation (FFD) based on B-splines. Normalized mutual information is used as a voxel-based similarity measure which is insensitive to intensity changes as a result of the contrast enhancement. Registration is achieved by minimizing a cost function, which represents a combination of the cost associated with the smoothness of the transformation and the cost associated with the image similarity. The algorithm has been applied to the fully automated registration of three-dimensional (3-D) breast MRI in volunteers and patients. In particular, we have compared the results of the proposed nonrigid registration algorithm to those obtained using rigid and affine registration techniques. The results clearly indicate that the nonrigid registration algorithm is much better able to recover the motion and deformation of the breast than rigid or affine registration algorithms.

All image-guided neurosurgical systems that we are aware of assume that the head and its contents behave as a rigid body. It is important to measure intraoperative brain deformation (brain shift) to provide some indication of the application accuracy of image-guided surgical systems, and also to provide data to develop and validate nonrigid registration algorithms to correct for such deformation. We are collecting data from patients undergoing neurosurgery in a high-field (1.5 T) interventional magnetic resonance (MR) scanner. High-contrast and high-resolution gradient-echo MR image volumes are collected immediately prior to surgery, during surgery, and at the end of surgery, with the patient intubated and lying on the operating table in the operative position. In this paper we report initial results from six patients: one freehand biopsy, one stereotactic functional procedure, and four resections. We investigate intraoperative brain deformation by examining threshold boundary overlays and difference images and by measuring ventricular volume. We also present preliminary results obtained using a nonrigid registration algorithm to quantify deformation. We found that some cases had much greater deformation than others, and also that, regardless of the procedure, there was very little deformation of the midline, the tentorium, the hemisphere contralateral to the procedure, and ipsilateral structures except those that are within 1 cm of the lesion or are gravitationally above the surgical site.

We present a new algorithm for the robust and accurate tracking of the aorta in cardiovascular magnetic resonance (MR) images. First, a rough estimate of the location and diameter of the aorta is obtained by applying a multiscale medial-response function using the available a priori knowledge. Then, this estimate is refined using an energy-minimizing deformable model which we define in a Markov-random-field (MRF) framework. In this context, we propose a global minimization technique based on stochastic relaxation, Simulated annealing (SA), which is shown to be superior to other minimization techniques, for minimizing the energy of the deformable model. We have evaluated the performance and robustness of the algorithm on clinical compliance studies in cardiovascular MR images. The segmentation and tracking has been successfully tested in spin-echo MR images of the aorta. The results show the ability of the algorithm to produce not only accurate, but also very reliable results in clinical routine applications.

We present a general formulation for a new knowledge-based approach to anisotropic diffusion of multi-feature and multi-dimensional images, with an illustrative application to cardiac MRI. We incorporate all available information through a more complete definition of the conductance function which differs from previous approaches in two aspects. First, we model the conductance as an explicit function of the position and not only of the differential geometry of the image data. Inherent properties of the system (such as geometrical features or non-homogeneous data sampling) can therefore be taken into account by allowing the conductance values to depend on the location in the spatial and temporal coordinate space. Secondly, by defining the conductance as a second rank tensor, the non-homogeneous diffusion equation gains a truly anisotropic character which is essential to emulate and handle certain aspects of complex data systems. We demonstrate the efficiency of the proposed framework ...

This paper presents a new segmentation algorithm by fitting active contour models (or snakes) to objects using adaptive splines. The adaptive spline model describes the contour of an object by a set of piecewisely interpolating C^2 polynomial spline patches which are locally controlled. Thus the resulting description of the object contour is continuous and smooth. Polynomial splines provide a fast and efficient way for interpolating the object contour and allow us to compute its internal energy due to bending and elasticity deformations analytically. The adaptive spline model can be represented by its spline control points. The accuracy of the model is gradually increased during the segmentation process by inserting new control points. For estimating the optimal position of the control points, two different relaxation techniques based on Markov Random Fields (MRFs) have been combined and evaluated: Simulated Annealing (SA), which is a stochastic relaxation technique, and Iterated Conditional Modes (ICM), which is a probabilistic relaxation technique. We have studied convergence behaviour and performance on artificial and medical images. The results show that the combination of both relaxation techniques provides very robust and initialization independent segmentation results.