Magnetic Resonance Imaging (MRI) suffers from long scan time which limits its clinical efficiency. MRI super-resolution and undersampling reconstruction alleviate this bottleneck by acquiring only partial signals and recovering the rest with generative algorithms. Besides, MRI can acquire volumes with different contrast weighting in one session. However, most methods for these tasks either operate on single-contrast image, ignoring cross-contrast structural consistency, or rely on expensive attention calculation for cross-contrast fusion with naive or over-complex structural awareness. We introduce SGF-MRI, a unified structure-guided framework that distills complementary structural information across contrasts to enhance both accelerated MRI reconstruction and super-resolution based on an efficient Multi-contrast Structural Distillation (MCSD) mechanism, specially designed to resolve the fundamental tradeoff between high-quality cross-contrast structural fusion and computational efficiency. Specifically, MCSD performs an adaptive cross-contrast neighbor similarity operation that is more lightweight than attention to extract structural information, which is then fed to an attention module to fuse with image features, enabling the model to efficiently attend to rich structural guidance in multi-contrast images. Our method achieves significantly higher computational efficiency and comparable restoration quality compared to state-of-the-art methods. Code will be provided upon acceptance.

Deformable multi-contrast image registration is a challenging yet crucial task due to the complex, non-linear intensity relationships across different imaging contrasts. Conventional registration methods typically rely on iterative optimization of the deformation field, which is time-consuming. Although recent learning-based approaches enable fast and accurate registration during inference, their generalizability remains limited to the specific contrasts observed during training. In this work, we propose an adaptive conditional contrast-agnostic deformable image registration framework (AC-CAR) based on a random convolution-based contrast augmentation scheme. AC-CAR can generalize to arbitrary imaging contrasts without observing them during training. To encourage contrast-invariant feature learning, we propose an adaptive conditional feature modulator (ACFM) that adaptively modulates the features and the contrast-invariant latent regularization to enforce the consistency of the learned feature across different imaging contrasts. Additionally, we enable our framework to provide contrast-agnostic registration uncertainty by integrating a variance network that leverages the contrast-agnostic registration encoder to improve the trustworthiness and reliability of AC-CAR. Experimental results demonstrate that AC-CAR outperforms baseline methods in registration accuracy and exhibits superior generalization to unseen imaging contrasts. Code is available at https://github.com/Yinsong0510/AC-CAR.

Cardiovascular health is vital to human well-being, and cardiac magnetic resonance (CMR) imaging is considered the clinical reference standard for diagnosing cardiovascular disease. However, its adoption is hindered by long scan times, complex contrasts, and inconsistent quality. While deep learning methods perform well on specific CMR imaging sequences, they often fail to generalize across modalities and sampling schemes. The lack of benchmarks for high-quality, fast CMR image reconstruction further limits technology comparison and adoption. The CMRxRecon2024 challenge, attracting over 200 teams from 18 countries, addressed these issues with two tasks: generalization to unseen modalities and robustness to diverse undersampling patterns. We introduced the largest public multi-modality CMR raw dataset, an open benchmarking platform, and shared code. Analysis of the best-performing solutions revealed that prompt-based adaptation and enhanced physics-driven consistency enabled strong cross-scenario performance. These findings establish principles for generalizable reconstruction models and advance clinically translatable AI in cardiovascular imaging.

Multimodal image-tabular learning is gaining attention, yet it faces challenges due to limited labeled data. While earlier work has applied self-supervised learning (SSL) to unlabeled data, its task-agnostic nature often results in learning suboptimal features for downstream tasks. Semi-supervised learning (SemiSL), which combines labeled and unlabeled data, offers a promising solution. However, existing multimodal SemiSL methods typically focus on unimodal or modality-shared features, ignoring valuable task-relevant modality-specific information, leading to a Modality Information Gap. In this paper, we propose STiL, a novel SemiSL tabular-image framework that addresses this gap by comprehensively exploring task-relevant information. STiL features a new disentangled contrastive consistency module to learn cross-modal invariant representations of shared information while retaining modality-specific information via disentanglement. We also propose a novel consensus-guided pseudo-labeling strategy to generate reliable pseudo-labels based on classifier consensus, along with a new prototype-guided label smoothing technique to refine pseudo-label quality with prototype embeddings, thereby enhancing task-relevant information learning in unlabeled data. Experiments on natural and medical image datasets show that STiL outperforms the state-of-the-art supervised/SSL/SemiSL image/multimodal approaches. Our code is available at https://github.com/siyi-wind/STiL.

In the face of the exponential growth of digital data, DNA is expected to become a new storage medium. Image data makes up a large proportion of digital data. However, existing DNA data storage models are mainly designed for general files. To address this issue, we propose a novel image encoding method for DNA data storage. We employ discrete wavelet transform to decompose the image and utilize an improved exponent-mantissa representation for numerical data. Subsequently, we achieve enhanced compression performance through context-adaptive arithmetic coding. Additionally, we construct a dictionary between ternary sequences and oligonucleotides to generate nucleotide sequences that meet the specified constraints. Experiments show that our method outperforms JPEG-DNA and BioCoder in compression performance and generates higher-quality nucleotide sequences.

Rapid head motion during sports collisions can cause traumatic brain injury. Head motion can be measured with instrumented mouthguards and fed into finite element (FE) models to predict brain strain, a measure of brain deformation and injury. Due to the computational cost of FE models, deep neural networks have been developed for near real-time prediction. However, they are not used in pitch-side assessments due to their complexity and reliance on full kinematic data, which cannot be reliably transmitted in real-time. We propose an extreme gradient boosting (XGBoost) model with simple input of two kinematic features. Its accuracy and efficiency were compared with two deep learning models: a multilayer perceptron (MLP) using 20 features, and a convolutional neural network (CNN) using entire kinematics. All models were trained on 1701 rugby impacts collected with mouthguards and simulated using the Imperial brain FE model. The XGBoost model predicted strain in key brain regions, while the deep learning models predicted whole-brain strain distributions. All models showed reasonable accuracy in predicting regional strain, with R 2 values 0.764–0.851 for XGBoost, 0.721–0.876 for MLP, and 0.744–0.887 for CNN. XGBoost required orders of magnitude fewer floating-point operations, and it used simple input that can be calculated on mouthguards and reliably transmitted in real-time. This study suggests that different models can be used at different stages of brain injury assessment. We hope that the XGBoost model proposed here will lower the barriers for adopting brain strain combined with instrumented mouthguards for pitch-side assessments from elite to grassroot collision sports.

White Matter Hyperintensities (WMH) are key neuroradiological markers of small vessel disease present in brain MRI. Assessment of WMH is important in research and clinics. However, WMH are challenging to segment due to their high variability in shape, location, size, poorly defined borders, and similar intensity profile to other pathologies (e.g stroke lesions) and artefacts (e.g head motion). In this work, we assess the utility and semantic properties of the most effective techniques for uncertainty quantification (UQ) in segmentation for the WMH segmentation task across multiple test-time data distributions. We find UQ techniques reduce 'silent failure' by identifying in UQ maps small WMH clusters in the deep white matter that are unsegmented by the model. A combination of Stochastic Segmentation Networks with Deep Ensembles also yields the highest Dice and lowest Absolute Volume Difference % (AVD) score and can highlight areas where there is ambiguity between WMH and stroke lesions. We further demonstrate the downstream utility of UQ, proposing a novel method for classification of the clinical Fazekas score using spatial features extracted from voxelwise WMH probability and UQ maps. We show that incorporating WMH uncertainty information improves Fazekas classification performance and calibration. Our model with (UQ and spatial WMH features)/(spatial WMH features)/(WMH volume only) achieves a balanced accuracy score of 0.74/0.67/0.62, and root brier score (↓) of 0.65/0.72/0.74 in the Deep WMH and balanced accuracy of 0.74/0.73/0.71 and root brier score of 0.64/0.66/0.68 in the Periventricular region. We further demonstrate that stochastic UQ techniques with high sample diversity can improve the detection of poor quality segmentations.

As an essential part of the future wireless networks, low earth orbit satellite networks (LEO-SN) is expected to achieve ubiquitous global networks access, in which the multibeam transmission is widely used to meet the increasing rate demand. However, in multibeam LEO-SN systems, the multiuser access and inter-group interference are crucial issues that need to be addressed urgently. To this end, we investigate the system weighted sum rate (WSR) maximization problem under the constraints of the user terminals (UTs) grouping, satellite total power budget, and minimum transmission rate requirements. For solving the problem, we propose a multiuser space-ground multicast group control (MU-SGMGC) scheme. Specifically, we first group all UTs into multiple multicast groups based on the channel correlation coefficient. Then, the group centers determination algorithm based on user distribution is proposed to ensure that each beam can cover all user groups. Finally, the beamformers designing problem is transformed into a difference-of-convex (DC) programming problem by utilizing auxiliary variables, and an iterative algorithm based on convex-concave procedure (CCP) is presented to solve the problem. Simulation results show that our proposed MU-SGMGC scheme has significant superiority in system WSR compared with the benchmark algorithms.

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.

This paper presents an effective and general data augmentation framework for medical image segmentation. We adopt a computationally efficient and data-efficient gradient-based meta-learning scheme to explicitly align the distribution of training and validation data which is used as a proxy for unseen test data. We improve the current data augmentation strategies with two core designs. First, we learn class-specific training-time data augmentation (TRA) effectively increasing the heterogeneity within the training subsets and tackling the class imbalance common in segmentation. Second, we jointly optimize TRA and test-time data augmentation (TEA), which are closely connected as both aim to align the training and test data distribution but were so far considered separately in previous works. We demonstrate the effectiveness of our method on four medical image segmentation tasks across different scenarios with two state-of-the-art segmentation models, DeepMedic and nnU-Net. Extensive experimentation shows that the proposed data augmentation framework can significantly and consistently improve the segmentation performance when compared to existing solutions. Code is publicly available at https://github.com/ZerojumpLine/JCSAugment.

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.

This paper investigates user grouping and beam-forming design in multi-beam low earth orbit (LEO) satellite communication (SATCOM) systems. To serve a great many user terminals (UTs) with a limited number of beams and improve the system performance, we formulate the weighted sum rate (WSR) maximization problem subject to the constraints of the UTs grouping, the satellite total power, and the minimum rate requirements of UTs. For solving this problem, we propose a multi-group multi-beamforming (MGMBF) scheme. In this scheme, all UTs are firstly adaptively grouped based on the channel correlation coefficients. Further, the beam centers are determined to ensure that all UTs are covered. After UTs grouping, slack variables are introduced to convert the beamforming design into a difference-of-convex (DC) programming problem. Moreover, an iterative algorithm is presented to solve the problem based on the convex-concave procedure (CCP), in which the beamforming vectors and slack variables are updated jointly by solving the convex sub-problem. Simulation results demonstrate that the MGMBF scheme improves the WSR by 25.1% compared with the MBIM algorithm, verifying the significant advantages of the proposed scheme.

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.

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.

Unsupervised domain adaptation challenges the problem of transferring knowledge from a well-labelled source domain to an unlabelled target domain. Recently, adversarial learning with bi-classifier has been proven effective in pushing cross-domain distributions close. Prior approaches typically leverage the disagreement between bi-classifier to learn transferable representations, however, they often neglect the classifier determinacy in the target domain, which could result in a lack of feature discriminability. In this paper, we present a simple yet effective method, namely Bi-Classifier Determinacy Maximization (BCDM), to tackle this problem. Motivated by the observation that target samples cannot always be separated distinctly by the decision boundary, here in the proposed BCDM, we design a novel classifier determinacy disparity (CDD) metric, which formulates classifier discrepancy as the class relevance of distinct target predictions and implicitly introduces constraint on the target feature discriminability. To this end, the BCDM can generate discriminative representations by encouraging target predictive outputs to be consistent and determined, meanwhile, preserve the diversity of predictions in an adversarial manner. Furthermore, the properties of CDD as well as the theoretical guarantees of BCDM's generalization bound are both elaborated. Extensive experiments show that BCDM compares favorably against the existing state-of-the-art domain adaptation methods.

Domain adaptation (DA) paves the way for label annotation and dataset bias issues by the knowledge transfer from a label-rich source domain to a related but unlabeled target domain. A mainstream of DA methods is to align the feature distributions of the two domains. However, the majority of them focus on the entire image features where irrelevant semantic information, e.g., the messy background, is inevitably embedded. Enforcing feature alignments in such case will negatively influence the correct matching of objects and consequently lead to the semantically negative transfer due to the confusion of irrelevant semantics. To tackle this issue, we propose Semantic Concentration for Domain Adaptation (SCDA), which encourages the model to concentrate on the most principal features via the pair-wise adversarial alignment of prediction distributions. Specifically, we train the classifier to class-wisely maximize the prediction distribution divergence of each sample pair, which enables the model to find the region with large differences among the same class of samples. Meanwhile, the feature extractor attempts to minimize that discrepancy, which suppresses the features of dissimilar regions among the same class of samples and accentuates the features of principal parts. As a general method, SCDA can be easily integrated into various DA methods as a regularizer to further boost their performance. Extensive experiments on the cross-domain benchmarks show the efficacy of SCDA.

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.

With the improving performance of microprocessors, their bandwidth and latency requirements have increased at a rate much faster than the network's ability to provide them. The performance bottleneck has gradually shifted from the processors to the interconnection network. Traditional electrical interconnection, because of its low bandwidth, high delay and low interconnection density, becomes one of the constraints of high-performance computer(HPC)'s further improvement. Due to the inherent limitations of the electrical interconnection network, optical interconnection has become a feasible way to improve the performance. A Three-Dimensional Optical Interconnection Network (TDOIN) with low latency and high throughput was put forward in our previous work. An accurate mathematical model can help us design and optimize the optical networks. However, there is not a mathematical model which can provide an accurate analysis for TDOIN. To make better use of the network, we propose a steady-state model to evaluate TDOIN's network characteristics in this paper. This mathematical model is constructed by using Probability Theory. The results of the model analysis and the simulation shows that our model can accurately depict behaviors of TDOIN at steady state.

In this paper, we propose a novel non-linear supervised metric learning algorithm. The algorithm combines the neighborhood component analysis method with constructive neural networks which gradually increase the network size during the training process. The network aims to maximize a stochastic variant of the leave-one-out K-nearest neighbor (KNN) score on the training set. In this way, the proposed algorithm learns a nonlinear metric for KNN classification, overcoming the limitations of traditional metric learning algorithms which are only capable of learning linear transformations. Therefore, the proposed method is more flexible and powerful in transforming data than its linear counterpart. Moreover, it can also learn a low-dimensional non-linear mapping for visualization and fast classification. We validate our method on several benchmark datasets both for metric learning and dimensionality reduction, and the results demonstrate the competitiveness of the proposed approach.