The rapid growth of large models' size has far outpaced that of computing resources. To bridge this gap, encouraged by the parsimonious relationship between genotype and phenotype in the brain's growth and development, we propose the so-called Hyper-Compression that turns the model compression into the issue of parameter representation via a hyperfunction. Specifically, it is known that the trajectory of some low-dimensional dynamic systems can fill the high-dimensional space eventually. Thus, Hyper-Compression, using these dynamic systems as the hyperfunctions, represents the parameters of the target network by their corresponding composition number or trajectory length. This suggests a novel mechanism for model compression, substantially different from the existing pruning, quantization, distillation, and decomposition. Along this direction, we methodologically identify a suitable dynamic system with the irrational winding as the hyperfunction and theoretically derive its associated error bound. Next, guided by our theoretical insights, we propose several engineering twists to make the Hyper-Compression pragmatic and effective. Lastly, systematic and comprehensive experiments on NLP models such as LLaMA and Qwen series and vision models confirm that Hyper-Compression enjoys the following PNAS merits: 1) Preferable compression ratio; 2) No post-hoc retraining; 3) Affordable inference time; and 4) Short compression time. It compresses LLaMA2-7B in an hour and achieves close-to-int4-quantization performance, without retraining and with a performance drop of less than 1%.

Self-supervised learning (SSL) has emerged as a crucial technique in image processing, encoding, and understanding, especially for developing today's vision foundation models that utilize large-scale datasets without annotations to enhance various downstream tasks. This study introduces a novel SSL approach, Information-Maximized Soft Variable Discretization (IMSVD), for image representation learning. Specifically, IMSVD softly discretizes each variable in the latent space, enabling the estimation of their probability distributions over training batches and allowing the learning process to be directly guided by information measures. Motivated by the MultiView assumption, we propose an information-theoretic objective function to learn transform-invariant, non-trivial, and redundancy-minimized representation features. We then derive a cross-joint entropy loss function for self-supervised image representation learning, which theoretically enjoys superiority over the existing methods in reducing feature redundancy. Notably, our non-contrastive IMSVD method statistically performs contrastive learning. Extensive experimental results demonstrate the effectiveness of IMSVD on various downstream tasks in terms of both accuracy and efficiency. Thanks to our variable discretization, the embedding features optimized by IMSVD offer unique explainability at the variable level. IMSVD has the potential to be adapted to other learning paradigms. Our code is publicly available at https://github.com/niuchuangnn/IMSVD.

The latest X-ray photon-counting computed tomography (PCCT) for extremity allows multi-energy high-resolution (HR) imaging for tissue characterization and material decomposition. However, both radiation dose and imaging speed need improvement for contrast-enhanced and other studies. Despite the success of deep learning methods for 2D few-view reconstruction, applying them to HR volumetric reconstruction of extremity scans for clinical diagnosis has been limited due to GPU memory constraints, training data scarcity, and domain gap issues. In this paper, we propose a deep learning-based approach for PCCT image reconstruction at halved dose and doubled speed in a New Zealand clinical trial. Particularly, we present a patch-based volumetric refinement network to alleviate the GPU memory limitation, train network with synthetic data, and use model-based iterative refinement to bridge the gap between synthetic and real-world data. The simulation and phantom experiments demonstrate consistently improved results under different acquisition conditions on both in- and off-domain structures using a fixed network. The image quality of 8 patients from the clinical trial are evaluated by three radiologists in comparison with the standard image reconstruction with a full-view dataset. It is shown that our proposed approach is essentially identical to or better than the clinical benchmark in terms of diagnostic image quality scores. Our approach has a great potential to improve the safety and efficiency of PCCT without compromising image quality.

Deep learning methods have impacted almost every research field, demonstrating notable successes in medical imaging tasks such as denoising and super-resolution. However, the prerequisite for deep learning is data at scale, but data sharing is expensive yet at risk of privacy leakage. As cutting-edge AI generative models, diffusion models have now become dominant because of their rigorous foundation and unprecedented outcomes. Here we propose a latent diffusion approach for data synthesis without compromising patient privacy. In our exemplary case studies, we develop a latent diffusion model to generate medical CT, MRI, and PET images using publicly available datasets. We demonstrate that state-of-the-art deep learning-based denoising/super-resolution networks can be trained on our synthetic data to achieve image quality with no significant difference from what the same network can achieve after being trained on the original data. In our advanced diffusion model, we specifically embed a safeguard mechanism to protect patient privacy effectively and efficiently. Our approach enables privacy-preserving public sharing of diverse big datasets for development of deep models, potentially enabling federated learning at the level of input data instead of local network weights.

Peer review is central to the integrity of scientific publishing. At IEEE Transactions on Medical Imaging (TMI), thousands of reviewers and editors work each year to ensure that accepted papers meet our high standards of significance, innovation, evaluation, and reproducibility (SIER) [1]. Yet the rapid growth in submissions, the increasing complexity of papers, and the decreasing availability of reviewers place mounting pressure on the TMI peer review system.

Computed tomography (CT) is a major medical imaging modality. Clinical CT scenarios, such as low-dose screening, sparse-view scanning, and metal implants, often lead to severe noise and artifacts in reconstructed images, requiring improved reconstruction techniques. The introduction of deep learning has significantly advanced CT image reconstruction. However, obtaining paired training data remains rather challenging due to patient motion and other constraints. Although deep learning methods can still perform well with approximately paired data, they inherently carry the risk of hallucination due to data inconsistencies and model instability. In this paper, we integrate the data fidelity with the state-of-the-art generative AI model, referred to as the Poisson flow generative model (PFGM) with a generalized version PFGM++, and propose a novel CT framework: Flow-Oriented Reconstruction Conditioning Engine (FORCE). In our experiments, the proposed method shows superior performance in various CT imaging tasks, outperforming existing unsupervised reconstruction approaches.

Datafication, the process where users’ actions online are pervasively recorded, tracked, aggregated, analysed, and exploited by online services in multiple ways, is becoming increasingly common today. However, we know little about how children, especially non-Western children, perceive such practices. Through one-to-one semi-structured interviews with 36 children aged 11–14 from Chinese middle schools, we examined how Chinese children perceive datafication practices. We identified three knowledge gaps in children’s current perceptions of datafication practices online, including their lack of recognition of (i) their data ownership, (ii) data being transmitted across platforms, and (iii) datafication could go beyond video recommendation and include inferences and profiling of their personal aspects. Through contextualising these observations within the Chinese context and its unique online ecosystem, we identified cultural traits in Chinese children’s perceptions of datafication. We drew on education theories to discuss how to support the future digital literacy development and design online platforms for Chinese children.

In computed tomography (CT) imaging, optimizing the balance between radiation dose and image quality is crucial due to the potentially harmful effects of radiation on patients. Although subjective assessments by radiologists are considered the gold standard in medical imaging, these evaluations can be time-consuming and costly. Thus, objective methods, such as the peak signal-to-noise ratio and structural similarity index measure, are often employed as alternatives. However, these metrics, initially developed for natural images, may not fully encapsulate the radiologists' assessment process. Consequently, interest in developing deep learning-based image quality assessment (IQA) methods that more closely align with radiologists' perceptions is growing. A significant barrier to this development has been the absence of open-source datasets and benchmark models specific to CT IQA. Addressing these challenges, we organized the Low-dose Computed Tomography Perceptual Image Quality Assessment Challenge in conjunction with the Medical Image Computing and Computer Assisted Intervention 2023. This event introduced the first open-source CT IQA dataset, consisting of 1,000 CT images of various quality, annotated with radiologists' assessment scores. As a benchmark, this challenge offers a comprehensive analysis of six submitted methods, providing valuable insight into their performance. This paper presents a summary of these methods and insights. This challenge underscores the potential for developing no-reference IQA methods that could exceed the capabilities of full-reference IQA methods, making a significant contribution to the research community with this novel dataset. The dataset is accessible at https://zenodo.org/records/7833096.

Cardiac computed tomography (CT) has emerged as a major imaging modality for the diagnosis and monitoring of cardiovascular diseases. High temporal resolution is essential to ensure diagnostic accuracy. Limited-angle data acquisition can reduce scan time and improve temporal resolution, but typically leads to severe image degradation and motivates for improved reconstruction techniques. In this paper, we propose a novel physics-informed score-based diffusion model (PSDM) for limited-angle reconstruction of cardiac CT. At the sampling time, we combine a data prior from a diffusion model and a model prior obtained via an iterative algorithm and Fourier fusion to further enhance the image quality. Specifically, our approach integrates the primal-dual hybrid gradient (PDHG) algorithm with score-based diffusion models, thereby enabling us to reconstruct high-quality cardiac CT images from limited-angle data. The numerical simulations and real data experiments confirm the effectiveness of our proposed approach.

X-ray computed tomography (CT) is widely used for medical diagnosis and treatment planning; however, concerns about ionizing radiation exposure drive efforts to optimize image quality at lower doses. This study introduces Poisson Flow Consistency Models (PFCM), a novel family of deep generative models that combines the robustness of PFGM++ with the efficient single-step sampling of consistency models. PFCM are derived by generalizing consistency distillation to PFGM++ through a change-of-variables and an updated noise distribution. As a distilled version of PFGM++, PFCM inherit the ability to trade off robustness for rigidity via the hyperparameter $\text {D} \in \text {(}{0},\infty \text {)}$ . A fact that we exploit to adapt this novel generative model for the task of low-dose CT image denoising, via a "task-specific" sampler that "hijacks" the generative process by replacing an intermediate state with the low-dose CT image. While this "hijacking" introduces a severe mismatch-the noise characteristics of low-dose CT images are different from that of intermediate states in the Poisson flow process-we show that the inherent robustness of PFCM at small D effectively mitigates this issue. The resulting sampler achieves excellent performance in terms of LPIPS, SSIM, and PSNR on the Mayo low-dose CT dataset. By contrast, an analogous sampler based on standard consistency models is found to be significantly less robust under the same conditions, highlighting the importance of a tunable D afforded by our novel framework. To highlight generalizability, we show effective denoising of clinical images from a prototype photon-counting system reconstructed using a sharper kernel and at a range of energy levels.

IEEE Transactions on Medical Imaging (TMI) publishes high-quality work that innovates imaging methods and advances medicine, science, and engineering. While artificial intelligence (AI) is currently prominent, the journal's scope extends well beyond AI-based imaging to encompass a full spectrum of imaging methods involving CT, MRI, PET, SPECT, ultrasound, optical, and hybrid systems, image reconstruction and processing (ranging from analytical and iterative algorithms to emerging deep imaging approaches), quantitative imaging and analysis (radiomics, biomarkers, and health analytics), image-guided interventions and therapy, as well as multimodal and multiscale imaging with integration of imaging and nonimaging data.

This editorial presents the vision and strategic direction of IEEE Transactions on Medical Imaging (TMI) under new leadership. Key points include restructuring the editorial board to enhance efficiency and diversity, streamlining the peer review process to improve decision quality and speed, and launching the AI for TMI (AI4TMI) initiative to integrate AI in journal management. Through these efforts, TMI aims to sustain excellence, adapt to emerging trends, and shape the future of medical imaging research.

Chest X-ray (CXR) is the most frequently ordered imaging test, supporting diverse clinical tasks from thoracic disease detection to postoperative monitoring. However, task-specific classification models are limited in scope, require costly labeled data, and lack generalizability to out-of-distribution datasets. To address these challenges, we introduce CheXFound, a self-supervised vision foundation model that learns robust CXR representations and generalizes effectively across a wide range of downstream tasks. We pretrained CheXFound on a curated CXR-987K dataset, comprising over approximately 987K unique CXRs from 12 publicly available sources. We propose a Global and Local Representations Integration (GLoRI) head for downstream adaptations, by incorporating fine- and coarse-grained disease-specific local features with global image features for enhanced performance in multilabel classification. Our experimental results showed that CheXFound outperformed state-of-the-art models in classifying 40 disease findings across different prevalence levels on the CXR-LT 24 dataset and exhibited superior label efficiency on downstream tasks with limited training data. Additionally, CheXFound achieved significant improvements on downstream tasks with out-of-distribution datasets, including opportunistic cardiovascular disease risk estimation, mortality prediction, malpositioned tube detection, and anatomical structure segmentation. The above results demonstrate CheXFound's strong generalization capabilities, which will enable diverse downstream adaptations with improved label efficiency in future applications. The project source code is publicly available at https://github.com/RPIDIAL/CheXFound.

Expertise scarcity and high cost of data annotation hinder the development of artificial intelligence (AI) foundation models for medical image analysis. Transfer learning provides a way to utilize the off-the-shelf foundation models to address the clinical challenges. However, such models encounter difficulties when adapting to new diseases not presented in their original pre-training datasets. Compounding this challenge is the limited availability of example cases for a new disease, which further leads to the poor performance of the existing transfer learning techniques. This paper proposes a novel method for transfer learning of foundation Vision-Language Models (VLMs) to efficiently adapt them to a new disease with only a few examples. Such an effective adaptation of VLMs hinges on learning the nuanced representation of new disease concepts. By capitalizing on the joint visual-linguistic capabilities of VLMs, we introduce disease-informed contextual prompting in a novel disease prototype learning framework, which enables VLMs to quickly grasp the concept of the new disease, even with limited data. Extensive experiments across multiple pre-trained medical VLMs and multiple tasks showcase the notable enhancements in performance compared to other existing adaptation techniques. The code will be made publicly available at https://github.com/RPIDIAL/Disease-informed-VLM-Adaptation.

Inspired by the diversity of biological neurons, quadratic artificial neurons can play an important role in deep learning models. The type of quadratic neurons of our interest replaces the inner-product operation in the conventional neuron with a quadratic function. Despite promising results so far achieved by networks of quadratic neurons, there are important issues not well addressed. Theoretically, the superior expressivity of a quadratic network over either a conventional network or a conventional network via quadratic activation is not fully elucidated, which makes the use of quadratic networks not well grounded. In practice, although a quadratic network can be trained via generic backpropagation, it can be subject to a higher risk of collapse than the conventional counterpart. To address these issues, we first apply the spline theory and a measure from algebraic geometry to give two theorems that demonstrate better model expressivity of a quadratic network than the conventional counterpart with or without quadratic activation. Then, we propose an effective training strategy referred to as referenced linear initialization (ReLinear) to stabilize the training process of a quadratic network, thereby unleashing the full potential in its associated machine learning tasks. Comprehensive experiments on popular datasets are performed to support our findings and confirm the performance of quadratic deep learning. We have shared our code in https://github.com/FengleiFan/ReLinear.

While various deep learning methods were proposed for low-dose computed tomography (CT) denoising, they often suffer from over-smoothing, blurring, and lack of explainability. To alleviate these issues, we propose a plug-and-play Language-Engaged Dual-space Alignment loss (LEDA) to optimize low-dose CT denoising models. Our idea is to leverage large language models (LLMs) to align denoised CT and normal-dose CT images in both the continuous perceptual space and discrete semantic space, which is the first LLM-based scheme for low-dose CT denoising. LEDA involves two steps: the first is to pretrain an LLM-guided CT autoencoder, which can encode a CT image into continuous high-level features and quantize them into a token space to produce semantic tokens derived from the LLM’s vocabulary; and the second is to minimize the discrepancy between the denoised CT images and normal-dose CT in terms of both encoded high-level features and quantized token embeddings derived by the LLM-guided CT autoencoder. Extensive experimental results demonstrate that our LEDA can enhance existing denoising models in terms of quantitative metrics and qualitative evaluation, and also provide explainability through language-level image understanding. The code is publicly available at https://github.com/hao1635/LEDA.

This article proposes a novel parallel management mode based on decentralized autonomous organizations (DAOs) for enterprises by utilizing the artificial systems, computational experiments, parallel execution (ACP) approach, parallel intelligence theory, and blockchain technologies, to realize the distributed management of an enterprise. The artificial enterprise DAO (EnDAO) corresponding to the actual enterprise is constructed, and they constitute a parallel system via virtual–real interaction and parallel execution. Through the non-fungible token (NFT)-based incentive mechanism, metaverse-based virtual learning and training, as well as DAO-based distributed management and decision-making, the management and control of the actual enterprise as well as its employees can be carried out. By virtue of the virtual–real interactions of three types of employees, as well as the virtual–real feedback of three closed loops in the parallel systems, DAO-based parallel management for enterprises can realize descriptive intelligence, predictive intelligence, and prescriptive intelligence. On this basis, this article takes the recruitment-oriented key performance indicator (KPI) management of a startup technology enterprise as the case to introduce the operation processes and illustrate the superiorities of the proposed DAO-based enterprise parallel management mode.

To break the dilemma of data island, a distributed framework for trustable control is proposed toward information security and data privacy in large-scale cyber-physical systems. The federated control system consists of distinct blockchains, as such information security and data privacy are technologically guaranteed. Moreover, data are divided into private and nonprivate data. Only nonprivate data can be exchanged for a better global system performance, where the interblockchain communication is ensured by cross-blockchain technologies. Federated control framework establishes a trustable environment where each subsystem is willing to share data for optimal performance. The architecture, structure, and implementation process of federated control are discussed, together with the potential applications to smart buildings.

This paper studies 3D low-dose computed tomography (CT) imaging. Although various deep learning methods were developed in this context, typically they focus on 2D images and perform denoising due to low-dose and deblurring for super-resolution separately. Up to date, little work was done for simultaneous in-plane denoising and through-plane deblurring, which is important to obtain high-quality 3D CT images with lower radiation and faster imaging speed. For this task, a straightforward method is to directly train an end-to-end 3D network. However, it demands much more training data and expensive computational costs. Here, we propose to link in-plane and through-plane transformers for simultaneous in-plane denoising and through-plane deblurring, termed as LIT-Former, which can efficiently synergize in-plane and through-plane sub-tasks for 3D CT imaging and enjoy the advantages of both convolution and transformer networks. LIT-Former has two novel designs: efficient multi-head self-attention modules (eMSM) and efficient convolutional feed-forward networks (eCFN). First, eMSM integrates in-plane 2D self-attention and through-plane 1D self-attention to efficiently capture global interactions of 3D self-attention, the core unit of transformer networks. Second, eCFN integrates 2D convolution and 1D convolution to extract local information of 3D convolution in the same fashion. As a result, the proposed LIT-Former synergizes these two sub-tasks, significantly reducing the computational complexity as compared to 3D counterparts and enabling rapid convergence. Extensive experimental results on simulated and clinical datasets demonstrate superior performance over state-of-the-art models. The source code is made available at https://github.com/hao1635/LIT-Former.

The presence of metal objects leads to corrupted CT projection measurements, resulting in metal artifacts in the reconstructed CT images. AI promises to offer improved solutions to estimate missing sinogram data for metal artifact reduction (MAR), as previously shown with convolutional neural networks (CNNs) and generative adversarial networks (GANs). Recently, denoising diffusion probabilistic models (DDPM) have shown great promise in image generation tasks, potentially outperforming GANs. In this study, a DDPM-based approach is proposed for inpainting of missing sinogram data for improved MAR. The proposed model is unconditionally trained, free from information on metal objects, which can potentially enhance its generalization capabilities across different types of metal implants compared to conditionally trained approaches. The performance of the proposed technique was evaluated and compared to the state-of-the-art normalized MAR (NMAR) approach as well as to CNN-based and GAN-based MAR approaches. The DDPM-based approach provided significantly higher SSIM and PSNR, as compared to NMAR (SSIM: p [Formula: see text]; PSNR: p [Formula: see text]), the CNN (SSIM: p [Formula: see text]; PSNR: p [Formula: see text]) and the GAN (SSIM: p [Formula: see text]; PSNR: p <0.05) methods. The DDPM-MAR technique was further evaluated based on clinically relevant image quality metrics on clinical CT images with virtually introduced metal objects and metal artifacts, demonstrating superior quality relative to the other three models. In general, the AI-based techniques showed improved MAR performance compared to the non-AI-based NMAR approach. The proposed methodology shows promise in enhancing the effectiveness of MAR, and therefore improving the diagnostic accuracy of CT.

Metal implants and other high-density objects in patients introduce severe streaking artifacts in CT images, compromising image quality and diagnostic performance. Although various methods were developed for CT metal artifact reduction over the past decades, including the latest dual-domain deep networks, remaining metal artifacts are still clinically challenging in many cases. Here we extend the state-of-the-art dual-domain deep network approach into a quad-domain counterpart so that all the features in the sinogram, image, and their corresponding Fourier domains are synergized to eliminate metal artifacts optimally without compromising structural subtleties. Our proposed quad-domain network for MAR, referred to as Quad-Net, takes little additional computational cost since the Fourier transform is highly efficient, and works across the four receptive fields to learn both global and local features as well as their relations. Specifically, we first design a Sinogram-Fourier Restoration Network (SFR-Net) in the sinogram domain and its Fourier space to faithfully inpaint metal-corrupted traces. Then, we couple SFR-Net with an Image-Fourier Refinement Network (IFR-Net) which takes both an image and its Fourier spectrum to improve a CT image reconstructed from the SFR-Net output using cross-domain contextual information. Quad-Net is trained on clinical datasets to minimize a composite loss function. Quad-Net does not require precise metal masks, which is of great importance in clinical practice. Our experimental results demonstrate the superiority of Quad-Net over the state-of-the-art MAR methods quantitatively, visually, and statistically. The Quad-Net code is publicly available at https://github.com/longzilicart/Quad-Net.

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.

Lowering radiation dose per view and utilizing sparse views per scan are two common CT scan modes, albeit often leading to distorted images characterized by noise and streak artifacts. Blind image quality assessment (BIQA) strives to evaluate perceptual quality in alignment with what radiologists perceive, which plays an important role in advancing low-dose CT reconstruction techniques. An intriguing direction involves developing BIQA methods that mimic the operational characteristic of the human visual system (HVS). The internal generative mechanism (IGM) theory reveals that the HVS actively deduces primary content to enhance comprehension. In this study, we introduce an innovative BIQA metric that emulates the active inference process of IGM. Initially, an active inference module, implemented as a denoising diffusion probabilistic model (DDPM), is constructed to anticipate the primary content. Then, the dissimilarity map is derived by assessing the interrelation between the distorted image and its primary content. Subsequently, the distorted image and dissimilarity map are combined into a multi-channel image, which is inputted into a transformer-based image quality evaluator. By leveraging the DDPM-derived primary content, our approach achieves competitive performance on a low-dose CT dataset.

The score-based generative model (SGM) has demonstrated remarkable performance in addressing challenging under-determined inverse problems in medical imaging. However, acquiring high-quality training datasets for these models remains a formidable task, especially in medical image reconstructions. Prevalent noise perturbations or artifacts in low-dose Computed Tomography (CT) or under-sampled Magnetic Resonance Imaging (MRI) hinder the accurate estimation of data distribution gradients, thereby compromising the overall performance of SGMs when trained with these data. To alleviate this issue, we propose a wavelet-improved denoising technique to cooperate with the SGMs, ensuring effective and stable training. Specifically, the proposed method integrates a wavelet sub-network and the standard SGM sub-network into a unified framework, effectively alleviating inaccurate distribution of the data distribution gradient and enhancing the overall stability. The mutual feedback mechanism between the wavelet sub-network and the SGM sub-network empowers the neural network to learn accurate scores even when handling noisy samples. This combination results in a framework that exhibits superior stability during the learning process, leading to the generation of more precise and reliable reconstructed images. During the reconstruction process, we further enhance the robustness and quality of the reconstructed images by incorporating regularization constraint. Our experiments, which encompass various scenarios of low-dose and sparse-view CT, as well as MRI with varying under-sampling rates and masks, demonstrate the effectiveness of the proposed method by significantly enhanced the quality of the reconstructed images. Especially, our method with noisy training samples achieves comparable results to those obtained using clean data. Our code at https://zenodo.org/record/8266123.

While classic studies proved that wide networks allow universal approximation, recent research and successes of deep learning demonstrate the power of deep networks. Based on a symmetric consideration, we investigate if the design of artificial neural networks should have a directional preference, and what the mechanism of interaction is between the width and depth of a network. Inspired by the De Morgan law, we address this fundamental question by establishing a quasi-equivalence between the width and depth of ReLU networks. We formulate two transforms for mapping an arbitrary ReLU network to a wide ReLU network and a deep ReLU network respectively, so that the essentially same capability of the original network can be implemented. Based on our findings, a deep network has a wide equivalent, and vice versa, subject to an arbitrarily small error.

To provide off-chain federations with complete privacy services to realize on-chain federated learning (FL), this article proposes a novel privacy framework for FL based on blockchain and smart contracts, named Artificial Identification. It consists of two modules: private peer-to-peer identification and private FL, using two scalable smart contracts to manage the identification and learning process, respectively. Based on Ethereum and interplenary file systems (IPFS), we implement our framework and comprehensively analyze its performance. Experiments show that the proposed framework has acceptable collaboration costs and offers advantages in terms of privacy, security, and decentralization. Furthermore, combined with radio frequency identification (RFID) technology, the framework has the potential to realize automatic on-chain identification and autonomous FL of machine clusters composed of Internet of Things (IoT) devices or distributed participants.

Crypto management is proposed to tackle the management decision-making challenges under data asymmetry and trust asymmetry that cannot be solved merely by technical means. It emphasizes the novel management model for the real-time generation of reliable, trustworthy, and usable management decisions based on blockchain and blockchain-driven technologies. First, the framework model of crypto management with detailed descriptions of each technique is introduced, where blockchain is the underlying technology, decentralized autonomous organization (DAO) is the management structure, federated data is the decision basis, smart contract is the decision method, and non-fungible token (NFT) is the main decision incentive. Then, its collaboration mechanisms of on-blockchain DAO and off-blockchain organization as well as intra-organization and extra-organization nodes are discussed. Moreover, the potential applications of crypto management are addressed, and a case of task-oriented performance management is given to state how crypto management works to generate the real-time management decisions. Toward the end, the future research directions are pointed out in this emerging new area.

Image denoising is a prerequisite for downstream tasks in many fields. Low-dose and photon-counting computed tomography (CT) denoising can optimize diagnostic performance at minimized radiation dose. Supervised deep denoising methods are popular but require paired clean or noisy samples that are often unavailable in practice. Limited by the independent noise assumption, current self-supervised denoising methods cannot process correlated noises as in CT images. Here we propose the first-of-its-kind similarity-based self-supervised deep denoising approach, referred to as Noise2Sim, that works in a nonlocal and nonlinear fashion to suppress not only independent but also correlated noises. Theoretically, Noise2Sim is asymptotically equivalent to supervised learning methods under mild conditions. Experimentally, Nosie2Sim recovers intrinsic features from noisy low-dose CT and photon-counting CT images as effectively as or even better than supervised learning methods on practical datasets visually, quantitatively and statistically. Noise2Sim is a general self-supervised denoising approach and has great potential in diverse applications.

In nuclear imaging, limited resolution causes partial volume effects (PVEs) that affect image sharpness and quantitative accuracy. Partial volume correction (PVC) methods incorporating high-resolution anatomical information from CT or MRI have been demonstrated to be effective. However, such anatomical-guided methods typically require tedious image registration and segmentation steps. Accurately segmented organ templates are also hard to obtain, particularly in cardiac SPECT imaging, due to the lack of hybrid SPECT/CT scanners with high-end CT and associated motion artifacts. Slight mis-registration/mis-segmentation would result in severe degradation in image quality after PVC. In this work, we develop a deep-learning-based method for fast cardiac SPECT PVC without anatomical information and associated organ segmentation. The proposed network involves a densely-connected multi-dimensional dynamic mechanism, allowing the convolutional kernels to be adapted based on the input images, even after the network is fully trained. Intramyocardial blood volume (IMBV) is introduced as an additional clinical-relevant loss function for network optimization. The proposed network demonstrated promising performance on 28 canine studies acquired on a GE Discovery NM/CT 570c dedicated cardiac SPECT scanner with a 64-slice CT using Technetium-99m-labeled red blood cells. This work showed that the proposed network with densely-connected dynamic mechanism produced superior results compared with the same network without such mechanism. Results also showed that the proposed network without anatomical information could produce images with statistically comparable IMBV measurements to the images generated by anatomical-guided PVC methods, which could be helpful in clinical translation.

Empowered by blockchain and Web3 technologies, decentralized autonomous organizations (DAOs) are able to redefine resources, production relations, and organizational structures in a revolutionary manner. This article aims to reanalyze DAOs from the perspectives of organization and operation, and provide a more precise definition of DAOs as Decentralized Autonomous Organizations and Operations. Based on this, the fundamental principles and requirements of DAOs are explained, while the infrastructure based on cyber–physical–social system (CPSS) and parallel intelligence, as well as the supporting technologies, such as digital twins, metaverse, and Web3, are discussed. Besides, a five-layer intelligent architecture is presented, and the closed-loop equation and new function-oriented intelligent algorithms are also proposed. Moreover, the governance mechanisms from the individual, organizational and social perspectives are discussed, and the incentive mechanisms for the human, robot, and digital human are analyzed. This article can be regarded as a stepping stone for further research and developments of DAOs.

In the context of Industrial Internet of Things (IIoT), RFID technologies have been widely applied to locate or track tagged objects for achieving item-level intelligence. However, prior localization work encounters two main issues. First, the phase measurement usually contains physical deviation. Existing localization work generally takes the physical center of an RFID antenna as its phase center, which is a key factor in improving localization accuracy but actually different from the physical center in practice. Second, the non-linear localization model is likely to be too complex to run on edge nodes with limited computing resources. In this paper, we present a LInear localizatiON solution, called LION, to perform the phase calibration for antennas with no need for the complex computation nor strong limitations. Specifically, we provide a novel lightweight model to pinpoint the actual antenna position quickly and accurately. Compared to previous localization methods, we reduce the intersection of circles or hyperbolas into radical lines, which greatly reduces the computation cost while guaranteeing the high accuracy. Further, to adapt to the complex environment with various ambient noise and multi-path effect, we leverage the weighted least square method to determine the optimal position. Moreover, we propose an adaptive parameter selection scheme to automatically choose optimal parameters for localization. In this way, LION is able to perform the accurate localization robustly. We implement LION using commercial RFID devices, and evaluate its performance extensively. Experimental results show the necessity of phase calibration as well as the high time efficiency of LION, e.g., the average accuracy improves by 6× and 2.1× for 2D and 3D localization, and the average time consuming is 0.02s and 1.8s for 2D and 3D cases.

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.

The similarity among samples and the discrepancy among clusters are two crucial aspects of image clustering. However, current deep clustering methods suffer from inaccurate estimation of either feature similarity or semantic discrepancy. In this paper, we present a Semantic Pseudo-labeling-based Image ClustEring (SPICE) framework, which divides the clustering network into a feature model for measuring the instance-level similarity and a clustering head for identifying the cluster-level discrepancy. We design two semantics-aware pseudo-labeling algorithms, prototype pseudo-labeling and reliable pseudo-labeling, which enable accurate and reliable self-supervision over clustering. Without using any ground-truth label, we optimize the clustering network in three stages: 1) train the feature model through contrastive learning to measure the instance similarity; 2) train the clustering head with the prototype pseudo-labeling algorithm to identify cluster semantics; and 3) jointly train the feature model and clustering head with the reliable pseudo-labeling algorithm to improve the clustering performance. Extensive experimental results demonstrate that SPICE achieves significant improvements (~10%) over existing methods and establishes the new state-of-the-art clustering results on six balanced benchmark datasets in terms of three popular metrics. Importantly, SPICE significantly reduces the gap between unsupervised and fully-supervised classification; e.g. there is only 2% (91.8% vs 93.8%) accuracy difference on CIFAR-10. Our code is made publicly available at https://github.com/niuchuangnn/SPICE.

Strategic traffic management is crucial for combating traffic congestion at the macroscopic level. However, such a field is still relatively unexplored, particularly for microscopic control objects, such as intersections and coordinated intersection groups. This article proposes a human-in-the-loop recommendation system for strategic urban traffic management, which follows an agent-based structure. A regional agent dispatcher is defined to assign agents for operation whenever “operation on-demand” is required. Such a requirement is identified by a daily-dependent operational mode on strategic traffic operations at a control object level. The strategic management scheme for each control object is guided by a strategic agent (customized), which is essentially a deep recommender model with a specific architecture. By featuring the multiagent design, a customized operational scheme can be generated at the intersection level, which instructs the corresponding controller to take specific operations. The utility of the recommendation system is demonstrated via a case study using real-world traffic data. In both offline and online evaluations, the system performs consistently at traffic operational recommendations in different scenarios and has the potential to provide more reasonable traffic operational strategies than a human-operated system.

Cine cardiac magnetic resonance imaging (MRI) is widely used for the diagnosis of cardiac diseases thanks to its ability to present cardiovascular features in excellent contrast. As compared to computed tomography (CT), MRI, however, requires a long scan time, which inevitably induces motion artifacts and causes patients' discomfort. Thus, there has been a strong clinical motivation to develop techniques to reduce both the scan time and motion artifacts. Given its successful applications in other medical imaging tasks such as MRI super-resolution and CT metal artifact reduction, deep learning is a promising approach for cardiac MRI motion artifact reduction. In this paper, we propose a novel recurrent generative adversarial network model for cardiac MRI motion artifact reduction. This model utilizes bi-directional convolutional long short-term memory (ConvLSTM) and multi-scale convolutions to improve the performance of the proposed network, in which bi-directional ConvLSTMs handle long-range temporal features while multi-scale convolutions gather both local and global features. We demonstrate a decent generalizability of the proposed method thanks to the novel architecture of our deep network that captures the essential relationship of cardiovascular dynamics. Indeed, our extensive experiments show that our method achieves better image quality for cine cardiac MRI images than existing state-of-the-art methods. In addition, our method can generate reliable missing intermediate frames based on their adjacent frames, improving the temporal resolution of cine cardiac MRI sequences.

Deep learning has attracted rapidly increasing attention in the field of tomographic image reconstruction, especially for CT, MRI, PET/SPECT, ultrasound and optical imaging. Among various topics, sparse-view CT remains a challenge which targets a decent image reconstruction from very few projections. To address this challenge, in this article we propose a Dual-domain Residual-based Optimization NEtwork (DRONE). DRONE consists of three modules respectively for embedding, refinement, and awareness. In the embedding module, a sparse sinogram is first extended. Then, sparse-view artifacts are effectively suppressed in the image domain. After that, the refinement module recovers image details in the residual data and image domains synergistically. Finally, the results from the embedding and refinement modules in the data and image domains are regularized for optimized image quality in the awareness module, which ensures the consistency between measurements and images with the kernel awareness of compressed sensing. The DRONE network is trained, validated, and tested on preclinical and clinical datasets, demonstrating its merits in edge preservation, feature recovery, and reconstruction accuracy.

Low-Dose CT (LDCT) can significantly improve the accuracy of lung cancer diagnosis and thus reduce cancer deaths compared to chest X-ray. The lung cancer risk population is also at high risk of other deadly diseases, for instance, cardiovascular diseases. Therefore, predicting the all-cause mortality risks of this population is of great importance. This paper introduces a knowledge-based analytical method using deep convolutional neural network (CNN) for all-cause mortality prediction. The underlying approach combines structural image features extracted from CNNs, based on LDCT volume at different scales, and clinical knowledge obtained from quantitative measurements, to predict the mortality risk of lung cancer screening subjects. The proposed method is referred as Knowledge-based Analysis of Mortality Prediction Network (KAMP-Net). It constitutes a collaborative framework that utilizes both imaging features and anatomical information, instead of completely relying on automatic feature extraction. Our work demonstrates the feasibility of incorporating quantitative clinical measurements to assist CNNs in all-cause mortality prediction from chest LDCT images. The results of this study confirm that radiologist defined features can complement CNNs in performance improvement. The experiments demonstrate that KAMP-Net can achieve a superior performance when compared to other methods.

Inspired by complexity and diversity of biological neurons, our group proposed quadratic neurons by replacing the inner product in current artificial neurons with a quadratic operation on input data, thereby enhancing the capability of an individual neuron. Along this direction, we are motivated to evaluate the power of quadratic neurons in popular network architectures, simulating human-like learning in the form of "quadratic-neuron-based deep learning". Our prior theoretical studies have shown important merits of quadratic neurons and networks in representation, efficiency, and interpretability. In this paper, we use quadratic neurons to construct an encoder-decoder structure, referred as the quadratic autoencoder, and apply it to low-dose CT denoising. The experimental results on the Mayo low-dose CT dataset demonstrate the utility and robustness of quadratic autoencoder in terms of image denoising and model efficiency. To our best knowledge, this is the first time that the deep learning approach is implemented with a new type of neurons and demonstrates a significant potential in the medical imaging field.

Magnetic resonance imaging (MRI) is widely used for screening, diagnosis, image-guided therapy, and scientific research. A significant advantage of MRI over other imaging modalities such as computed tomography (CT) and nuclear imaging is that it clearly shows soft tissues in multi-contrasts. Compared with other medical image super-resolution methods that are in a single contrast, multi-contrast super-resolution studies can synergize multiple contrast images to achieve better super-resolution results. In this paper, we propose a one-level non-progressive neural network for low up-sampling multi-contrast super-resolution and a two-level progressive network for high up-sampling multi-contrast super-resolution. The proposed networks integrate multi-contrast information in a high-level feature space and optimize the imaging performance by minimizing a composite loss function, which includes mean-squared-error, adversarial loss, perceptual loss, and textural loss. Our experimental results demonstrate that 1) the proposed networks can produce MRI super-resolution images with good image quality and outperform other multi-contrast super-resolution methods in terms of structural similarity and peak signal-to-noise ratio; 2) combining multi-contrast information in a high-level feature space leads to a significantly improved result than a combination in the low-level pixel space; and 3) the progressive network produces a better super-resolution image quality than the non-progressive network, even if the original low-resolution images were highly down-sampled.

In this paper, we present a semi-supervised deep learning approach to accurately recover high-resolution (HR) CT images from low-resolution (LR) counterparts. Specifically, with the generative adversarial network (GAN) as the building block, we enforce the cycle-consistency in terms of the Wasserstein distance to establish a nonlinear end-to-end mapping from noisy LR input images to denoised and deblurred HR outputs. We also include the joint constraints in the loss function to facilitate structural preservation. In this process, we incorporate deep convolutional neural network (CNN), residual learning, and network in network techniques for feature extraction and restoration. In contrast to the current trend of increasing network depth and complexity to boost the imaging performance, we apply a parallel 1×1 CNN to compress the output of the hidden layer and optimize the number of layers and the number of filters for each convolutional layer. The quantitative and qualitative evaluative results demonstrate that our proposed model is accurate, efficient and robust for super-resolution (SR) image restoration from noisy LR input images. In particular, we validate our composite SR networks on three large-scale CT datasets, and obtain promising results as compared to the other state-of-the-art methods.

Noise and artifacts are intrinsic to low-dose computed tomography (LDCT) data acquisition, and will significantly affect the imaging performance. Perfect noise removal and image restoration is intractable in the context of LDCT due to the statistical and the technical uncertainties. In this letter, we apply the generative adversarial network (GAN) framework with a visual attention mechanism to deal with this problem in a data-driven/machine learning fashion. Our main idea is to inject visual attention knowledge into the learning process of GAN to provide a powerful prior of the noise distribution. By doing this, both the generator and discriminator networks are empowered with visual attention information so that they will not only pay special attention to noisy regions and surrounding structures but also explicitly assess the local consistency of the recovered regions. Our experiments qualitatively and quantitatively demonstrate the effectiveness of the proposed method with clinic CT images.

Compressive sensing (CS) has proved effective for tomographic reconstruction from sparsely collected data or under-sampled measurements, which are practically important for few-view computed tomography (CT), tomosynthesis, interior tomography, and so on. To perform sparse-data CT, the iterative reconstruction commonly uses regularizers in the CS framework. Currently, how to choose the parameters adaptively for regularization is a major open problem. In this paper, inspired by the idea of machine learning especially deep learning, we unfold the state-of-the-art "fields of experts"-based iterative reconstruction scheme up to a number of iterations for data-driven training, construct a learned experts' assessment-based reconstruction network (LEARN) for sparse-data CT, and demonstrate the feasibility and merits of our LEARN network. The experimental results with our proposed LEARN network produces a superior performance with the well-known Mayo Clinic low-dose challenge data set relative to the several state-of-the-art methods, in terms of artifact reduction, feature preservation, and computational speed. This is consistent to our insight that because all the regularization terms and parameters used in the iterative reconstruction are now learned from the training data, our LEARN network utilizes application-oriented knowledge more effectively and recovers underlying images more favorably than competing algorithms. Also, the number of layers in the LEARN network is only 50, reducing the computational complexity of typical iterative algorithms by orders of magnitude.

Low-dose computed tomography (LDCT) has attracted major attention in the medical imaging field, since CT-associated X-ray radiation carries health risks for patients. The reduction of the CT radiation dose, however, compromises the signal-to-noise ratio, which affects image quality and diagnostic performance. Recently, deep-learning-based algorithms have achieved promising results in LDCT denoising, especially convolutional neural network (CNN) and generative adversarial network (GAN) architectures. This paper introduces a conveying path-based convolutional encoder-decoder (CPCE) network in 2-D and 3-D configurations within the GAN framework for LDCT denoising. A novel feature of this approach is that an initial 3-D CPCE denoising model can be directly obtained by extending a trained 2-D CNN, which is then fine-tuned to incorporate 3-D spatial information from adjacent slices. Based on the transfer learning from 2-D to 3-D, the 3-D network converges faster and achieves a better denoising performance when compared with a training from scratch. By comparing the CPCE network with recently published work based on the simulated Mayo data set and the real MGH data set, we demonstrate that the 3-D CPCE denoising model has a better performance in that it suppresses image noise and preserves subtle structures.

In[1], please note the updated figure captions for Figures 5, 6, 7, and 8 as follows:

Over past several years, machine learning, or more generally artificial intelligence, has generated overwhelming research interest and attracted unprecedented public attention. As tomographic imaging researchers, we share the excitement from our imaging perspective [item 1) in the Appendix], and organized this special issue dedicated to the theme of "Machine learning for image reconstruction." This special issue is a sister issue of the special issue published in May 2016 of this journal with the theme "Deep learning in medical imaging" [item 2) in the Appendix]. While the previous special issue targeted medical image processing/analysis, this special issue focuses on data-driven tomographic reconstruction. These two special issues are highly complementary, since image reconstruction and image analysis are two of the main pillars for medical imaging. Together we cover the whole workflow of medical imaging: from tomographic raw data/features to reconstructed images and then extracted diagnostic features/readings.

The continuous development and extensive use of computed tomography (CT) in medical practice has raised a public concern over the associated radiation dose to the patient. Reducing the radiation dose may lead to increased noise and artifacts, which can adversely affect the radiologists' judgment and confidence. Hence, advanced image reconstruction from low-dose CT data is needed to improve the diagnostic performance, which is a challenging problem due to its ill-posed nature. Over the past years, various low-dose CT methods have produced impressive results. However, most of the algorithms developed for this application, including the recently popularized deep learning techniques, aim for minimizing the mean-squared error (MSE) between a denoised CT image and the ground truth under generic penalties. Although the peak signal-to-noise ratio is improved, MSE- or weighted-MSE-based methods can compromise the visibility of important structural details after aggressive denoising. This paper introduces a new CT image denoising method based on the generative adversarial network (GAN) with Wasserstein distance and perceptual similarity. The Wasserstein distance is a key concept of the optimal transport theory and promises to improve the performance of GAN. The perceptual loss suppresses noise by comparing the perceptual features of a denoised output against those of the ground truth in an established feature space, while the GAN focuses more on migrating the data noise distribution from strong to weak statistically. Therefore, our proposed method transfers our knowledge of visual perception to the image denoising task and is capable of not only reducing the image noise level but also trying to keep the critical information at the same time. Promising results have been obtained in our experiments with clinical CT images.

Despite the rapid developments of X-ray cone-beam CT (CBCT), image noise still remains a major issue for the low dose CBCT. To suppress the noise effectively while retain the structures well for low dose CBCT image, in this paper, a sparse constraint based on the 3-D dictionary is incorporated into a regularized iterative reconstruction framework, defining the 3-D dictionary learning (3-DDL) method. In addition, by analyzing the sparsity level curve associated with different regularization parameters, a new adaptive parameter selection strategy is proposed to facilitate our 3-DDL method. To justify the proposed method, we first analyze the distributions of the representation coefficients associated with the 3-D dictionary and the conventional 2-D dictionary to compare their efficiencies in representing volumetric images. Then, multiple real data experiments are conducted for performance validation. Based on these results, we found: 1) the 3-D dictionary-based sparse coefficients have three orders narrower Laplacian distribution compared with the 2-D dictionary, suggesting the higher representation efficiencies of the 3-D dictionary; 2) the sparsity level curve demonstrates a clear Z-shape, and hence referred to as Z-curve, in this paper; 3) the parameter associated with the maximum curvature point of the Z-curve suggests a nice parameter choice, which could be adaptively located with the proposed Z-index parameterization (ZIP) method; 4) the proposed 3-DDL algorithm equipped with the ZIP method could deliver reconstructions with the lowest root mean squared errors and the highest structural similarity index compared with the competing methods; 5) similar noise performance as the regular dose FDK reconstruction regarding the standard deviation metric could be achieved with the proposed method using (1/2)/(1/4)/(1/8) dose level projections. The contrast-noise ratio is improved by ~2.5/3.5 times with respect to two different cases under the (1/8) dose level compared with the low dose FDK reconstruction. The proposed method is expected to reduce the radiation dose by a factor of 8 for CBCT, considering the voted strongly discriminated low contrast tissues.

Given the potential risk of X-ray radiation to the patient, low-dose CT has attracted a considerable interest in the medical imaging field. Currently, the main stream low-dose CT methods include vendor-specific sinogram domain filtration and iterative reconstruction algorithms, but they need to access raw data, whose formats are not transparent to most users. Due to the difficulty of modeling the statistical characteristics in the image domain, the existing methods for directly processing reconstructed images cannot eliminate image noise very well while keeping structural details. Inspired by the idea of deep learning, here we combine the autoencoder, deconvolution network, and shortcut connections into the residual encoder-decoder convolutional neural network (RED-CNN) for low-dose CT imaging. After patch-based training, the proposed RED-CNN achieves a competitive performance relative to the-state-of-art methods in both simulated and clinical cases. Especially, our method has been favorably evaluated in terms of noise suppression, structural preservation, and lesion detection.

Spectral computed tomography (CT) produces an energy-discriminative attenuation map of an object, extending a conventional image volume with a spectral dimension. In spectral CT, an image can be sparsely represented in each of multiple energy channels, and are highly correlated among energy channels. According to this characteristics, we propose a tensor-based dictionary learning method for spectral CT reconstruction. In our method, tensor patches are extracted from an image tensor, which is reconstructed using the filtered backprojection (FBP), to form a training dataset. With the Candecomp/Parafac decomposition, a tensor-based dictionary is trained, in which each atom is a rank-one tensor. Then, the trained dictionary is used to sparsely represent image tensor patches during an iterative reconstruction process, and the alternating minimization scheme is adapted for optimization. The effectiveness of our proposed method is validated with both numerically simulated and real preclinical mouse datasets. The results demonstrate that the proposed tensor-based method generally produces superior image quality, and leads to more accurate material decomposition than the currently popular popular methods.

Non-rigid registration finds many applications such as photogrammetry, motion tracking, model retrieval, and object recognition. In this paper we propose a novel convex hull aided registration method (CHARM) to match two point sets subject to a non-rigid transformation. First, two convex hulls are extracted from the source and target respectively. Then, all points of the point sets are projected onto the reference plane through each triangular facet of the hulls. From these projections, invariant features are extracted and matched optimally. The matched feature point pairs are mapped back onto the triangular facets of the convex hulls to remove outliers that are outside any relevant triangular facet. The rigid transformation from the source to the target is robustly estimated by the random sample consensus (RANSAC) scheme through minimizing the distance between the matched feature point pairs. Finally, these feature points are utilized as the control points to achieve non-rigid deformation in the form of thin-plate spline of the entire source point set towards the target one. The experimental results based on both synthetic and real data show that the proposed algorithm outperforms several state-of-the-art ones with respect to sampling, rotational angle, and data noise. In addition, the proposed CHARM algorithm also shows higher computational efficiency compared to these methods.

Spectral CT with photon counting detectors can significantly improve CT performance by reducing image noise and dose, increasing contrast resolution and material specificity, as well as enabling functional and molecular imaging with existing and emerging probes. However, the current photon counting detector architecture is difficult to balance the number of energy bins and the statistical noise in each energy bin. Moreover, the hardware support for multi-energy bins demands a complex circuit which is expensive. In this paper, we promote a new scheme known as hybrid detectors that combine the dynamic-threshold-based counting and integrating modes. In this scheme, an energy threshold can be dynamically changed during a spectral CT scan, which can be considered as compressive sensing along the spectral dimension. By doing so, the number of energy bins can be retrospectively specified, even in a spatially varying fashion. To establish the feasibility and merits of such hybrid detectors, we develop a tensor-based PRISM algorithm to reconstruct a spectral CT image from dynamic dual-energy data, and perform experiments with simulated and real data, producing very promising results.

This special issue serves as a forum of high visibility and synergy to promote the momentum of spectral CT. Through a rigorous peer-review process, 14 high-quality papers1 have been included from leading groups around the world. These papers give a panorama of the state of the art, addressing challenges in detector and source technologies, image reconstruction, material decomposition, performance evaluation, biomedical, and other applications.

The long-standing interior problem has important mathematical and practical implications. The recently developed interior tomography methods have produced encouraging results. A particular scenario for theoretically exact interior reconstruction from truncated projections is that there is a known sub-region in the ROI. In this paper, we improve a novel continuous singular value decomposition (SVD) method for interior reconstruction assuming a known sub-region. First, two sets of orthogonal eigen-functions are calculated for the Hilbert and image spaces respectively. Then, after the interior Hilbert data are calculated from projection data through the ROI, they are projected onto the eigen-functions in the Hilbert space, and an interior image is recovered by a linear combination of the eigen-functions with the resulting coefficients. Finally, the interior image is compensated for the ambiguity due to the null space utilizing the prior sub-region knowledge. Experiments with simulated and real data demonstrate the advantages of our approach relative to the POCS type interior reconstructions.

Although diagnostic medical imaging provides enormous benefits in the early detection and accuracy diagnosis of various diseases, there are growing concerns on the potential side effect of radiation induced genetic, cancerous and other diseases. How to reduce radiation dose while maintaining the diagnostic performance is a major challenge in the computed tomography (CT) field. Inspired by the compressive sensing theory, the sparse constraint in terms of total variation (TV) minimization has already led to promising results for low-dose CT reconstruction. Compared to the discrete gradient transform used in the TV method, dictionary learning is proven to be an effective way for sparse representation. On the other hand, it is important to consider the statistical property of projection data in the low-dose CT case. Recently, we have developed a dictionary learning based approach for low-dose X-ray CT. In this paper, we present this method in detail and evaluate it in experiments. In our method, the sparse constraint in terms of a redundant dictionary is incorporated into an objective function in a statistical iterative reconstruction framework. The dictionary can be either predetermined before an image reconstruction task or adaptively defined during the reconstruction process. An alternating minimization scheme is developed to minimize the objective function. Our approach is evaluated with low-dose X-ray projections collected in animal and human CT studies, and the improvement associated with dictionary learning is quantified relative to filtered backprojection and TV-based reconstructions. The results show that the proposed approach might produce better images with lower noise and more detailed structural features in our selected cases. However, there is no proof that this is true for all kinds of structures.

SUMMARY: In vivo dynamic contrast-enhanced imaging tools provide non-invasive methods for analyzing various functional changes associated with disease initiation, progression and responses to therapy. The quantitative application of these tools has been hindered by its inability to accurately resolve and characterize targeted tissues due to spatially mixed tissue heterogeneity. Convex Analysis of Mixtures - Compartment Modeling (CAM-CM) signal deconvolution tool has been developed to automatically identify pure-volume pixels located at the corners of the clustered pixel time series scatter simplex and subsequently estimate tissue-specific pharmacokinetic parameters. CAM-CM can dissect complex tissues into regions with differential tracer kinetics at pixel-wise resolution and provide a systems biology tool for defining imaging signatures predictive of phenotypes. AVAILABILITY: The MATLAB source code can be downloaded at the authors' website www.cbil.ece.vt.edu/software.htm CONTACT: [email protected] SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

We clarify that the derivative-free fan-beam reconstruction formula [IEEE Trans. Image Process. 2, 543-547, 1993] only allows exact reconstruction of an object for a circular trajectory or at the origin of the coordinate system for a radially symmetric noncircular trajectory.

Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) provides a noninvasive method for evaluating tumor vasculature patterns based on contrast accumulation and washout. However, due to limited imaging resolution and tumor tissue heterogeneity, tracer concentrations at many pixels often represent a mixture of more than one distinct compartment. This pixel-wise partial volume effect (PVE) would have profound impact on the accuracy of pharmacokinetics studies using existing compartmental modeling (CM) methods. We, therefore, propose a convex analysis of mixtures (CAM) algorithm to explicitly mitigate PVE by expressing the kinetics in each pixel as a nonnegative combination of underlying compartments and subsequently identifying pure volume pixels at the corners of the clustered pixel time series scatter plot simplex. The algorithm is supported theoretically by a well-grounded mathematical framework and practically by plug-in noise filtering and normalization preprocessing. We demonstrate the principle and feasibility of the CAM-CM approach on realistic synthetic data involving two functional tissue compartments, and compare the accuracy of parameter estimates obtained with and without PVE elimination using CAM or other relevant techniques. Experimental results show that CAM-CM achieves a significant improvement in the accuracy of kinetic parameter estimation. We apply the algorithm to real DCE-MRI breast cancer data and observe improved pharmacokinetic parameter estimation, separating tumor tissue into regions with differential tracer kinetics on a pixel-by-pixel basis and revealing biologically plausible tumor tissue heterogeneity patterns. This method combines the advantages of multivariate clustering, convex geometry analysis, and compartmental modeling approaches. The open-source MATLAB software of CAM-CM is publicly available from the Web.

Compressive sensing (CS) has seen impressive successes and fast growth over the past ten years, including applications in medical imaging. Applications of CS to magnetic resonance imaging (MRI) have been the earliest, most numerous, and most diverse, owing to the tremendous flexibility in designing the acquisition process and the pressing need that MRI has, as a slow acquisition modality, to reduce the sampling requirements.

This paper presents a statistical interior tomography (SIT) approach making use of compressed sensing (CS) theory. With the projection data modeled by the Poisson distribution, an objective function with a total variation (TV) regularization term is formulated in the maximization of a posteriori (MAP) framework to solve the interior problem. An alternating minimization method is used to optimize the objective function with an initial image from the direct inversion of the truncated Hilbert transform. The proposed SIT approach is extensively evaluated with both numerical and real datasets. The results demonstrate that SIT is robust with respect to data noise and down-sampling, and has better resolution and less bias than its deterministic counterpart in the case of low count data.

Cardiac computed tomography (CT) has been improved over past years, but it still needs improvement for higher temporal resolution in the cases of high or irregular cardiac rates. Given successful applications of dual-source cardiac CT scanners, triple-source cone-beam CT seems a promising mode for cardiac CT. In this paper, we propose two filtered-backprojection algorithms for triple-source helical cone-beam CT. The first algorithm utilizes two families of filtering lines. These lines are parallel to the tangent of the scanning trajectory and the so-called L lines. The second algorithm utilizes two families of filtering lines tangent to the boundaries of the Zhao window and L lines, respectively, but it eliminates the filtering paths along the tangent of the scanning trajectory, thus reducing the required detector size greatly. The first algorithm is theoretically exact for r < 0.265R and quasi-exact for 0.265R <or= r < 0.495R, and the second algorithm is quasi-exact for r < 0.495R , where r and R denote the object radius and the trajectory radius, respectively. Both algorithms are computationally efficient. Numerical results are presented to verify and showcase the proposed algorithms.

Multisource cone-beam computed tomography (CT) is an attractive approach of choice for superior temporal resolution, which is critically important for cardiac imaging and contrast enhanced studies. In this paper, we present a filtered-backprojection (FBP) algorithm for triple-source helical cone-beam CT. The algorithm is both exact and efficient. It utilizes data from three inter-helix PI-arcs associated with the inter-helix PI-lines and the minimum detection windows defined for the triple-source configuration. The proof of the formula is based on the geometric relations specific to triple-source helical cone-beam scanning. Simulation results demonstrate the validity of the reconstruction algorithm. This algorithm is also extended to a multisource version for (2N + 1)-source helical cone-beam CT. With parallel computing, the proposed FBP algorithms can be significantly faster than our previously published multisource backprojection-filtration algorithms. Thus, the FBP algorithms are promising in applications of triple-source helical cone-beam CT.

It is well known that a rigid in-plane motion can be decomposed into a translation and a rotation around an origin. Based on our previous work, we first extend the Helgason-Ludwig consistency condition (HLCC) to cover a general rigid motion in fan-beam geometry. Then, we model the general motion by several parameters, and develop an iterative scheme for estimation of the in-plane motion parameters. This scheme determines the motion parameters by numerically minimizing an objective function constructed based on the HLCC. After the motion parameters are estimated, image reconstruction can be performed to compensate for the motion effects. Finally, we implement the algorithm and evaluate its performance in numerical simulations.

Using the group theory, we formulate integral invariants of projection data in fan-beam and cone-beam computed tomography (CT), which can be applied to sense an object motion and detect a contrast bolus arrival

A basic assumption in the classic computed tomography (CT) theory is that an object remains stationary in an entire scan. In biomedical CT/micro-CT, this assumption is often violated. To produce high-resolution images, such as for our recently proposed clinical micro-CT (CMCT) prototype, it is desirable to develop a precise motion estimation and image reconstruction scheme. In this paper, we first extend the Helgason-Ludwig consistency condition (HLCC) from parallel-beam to fan-beam geometry when an object is subject to a translation. Then, we propose a novel method to estimate the motion parameters only from sinograms based on the HLCC. To reconstruct the moving object, we formulate two generalized fan-beam reconstruction methods, which are in filtered backprojection and backprojection filtering formats, respectively. Furthermore, we present numerical simulation results to show that our approach is accurate and robust.

In this letter, we present a new fan-beam CT formula, based on which we discuss the relation between the filtered backprojection (FBP) algorithm and the backprojection (BP) algorithm. Specifically, the FBP algorithm can be expressed in a series with its first-order approximation being the BP algorithm. As a result, we identify a link between X-ray CT and number theory.

While ultrasound- and magnetic resonance-based elastography techniques have proved to be powerful biomedical imaging tools, most approaches assume isotropic material properties. In this paper, a general framework is developed for tomography-based anisotropic elastography. An anatomically well-motivated piece-wise homogeneous model is proposed to represent a class of biological objects consisting of different regions. With established tomography modality, static displacements are measured on the entire external and internal boundaries, and the force distribution is recorded on part of the external surface. A principle is proposed to identify the anisotropic elastic moduli of the constituent regions with the obtained boundary measurements. The reconstruction procedure is optimization-based with minimizing an objective function that measures the difference between the predicted and observed displacements. Analytic gradients of the objective function with respect to the elastic moduli are calculated using an adjoint method, and are utilized to significantly improve the numerical efficiency. Simulations are performed to identify the elastic moduli in a breast phantom consisting of soft tissue and a hard tumor. For isotropic phantom, one set of the boundary measurements enables unique reconstruction results for the tissue and tumor. For anisotropic phantom, however, multiple sets of the measurements corresponding to different deformation modes become necessary.

In this paper, we prove a generalized backprojection-filtration formula for exact cone-beam image reconstruction with an arbitrary scanning locus. Our proof is independent of the shape of the scanning locus, as long as the object is contained in a region where there is a chord through any interior point. As special cases, this generalized formula can be applied with cone-beam scanning along nonstandard spiral and saddle curves, as well as in an n-PI window setting. The algorithmic implementation and numerical results are described to support the correctness of our general claim.

Computed tomography (CT) has been extensively studied for years and widely used in the modern society. Although the filtered back-projection algorithm is the method of choice by manufacturers, efforts are being made to revisit iterative methods due to their unique advantages, such as superior performance with incomplete noisy data. In 1984, the simultaneous algebraic reconstruction technique (SART) was developed as a major refinement of the algebraic reconstruction technique (ART). However, the convergence of the SART has never been established since then. In this paper, the convergence is proved under the condition that coefficients of the linear imaging system are nonnegative. It is shown that from any initial guess the sequence generated by the SART converges to a weighted least square solution.

We introduce a general iterative scheme for image reconstruction based on Landweber's method. In our configuration, a sequential block-iterative (SeqBI) version can be readily formulated from a simultaneous block-iterative (SimBI) version, and vice versa. This provides a mechanism to derive new algorithms from known ones. It is shown that some widely used iterative algorithms, such as the algebraic reconstruction technique (ART), simultaneous ART (SART), Cimmino's, and the recently designed diagonal weighting and component averaging algorithms, are special examples of the general scheme. We prove convergence of the general scheme under conditions more general than assumed in earlier studies, for its SeqBI and SimBI versions in the consistent and inconsistent cases, respectively. Our results suggest automatic relaxation strategies for the SeqBI and SimBI versions and characterize the dependence of the limit image on the initial guess. It is found that in all cases the limit is the sum of the minimum norm solution of a weighted least-squares problem and an oblique projection of the initial image onto the null space of the system matrix.

To discriminate fine anatomical features in the inner ear, it has been desirable that spiral computed tomography (CT) may perform beyond their current resolution limits with the aid of digital image processing techniques. In this paper, we develop a blind deblurring approach to enhance image resolution retrospectively without complete knowledge of the underlying point spread function (PSF). An oblique CT image can be approximated as the convolution of an isotropic Gaussian PSF and the actual cross section. Practically, the parameter of the PSF is often unavailable. Hence, estimation of the parameter for the underlying PSF is crucially important for blind image deblurring. Based on the iterative deblurring theory, we formulate an edge-to-noise ratio (ENR) to characterize the image quality change due to deblurring. Our blind deblurring algorithm estimates the parameter of the PSF by maximizing the ENR, and deblurs images. In the phantom studies, the blind deblurring algorithm reduces image blurring by about 24%, according to our blurring residual measure. Also, the blind deblurring algorithm works well in patient studies. After fully automatic blind deblurring, the conspicuity of the submillimeter features of the cochlea is substantially improved.

We generalize the axiomatic quantification of one-dimensional (1-D) image resolution to the multidimensional case. The imaging system of interest is characterized by a nonnegative spatially invariant point spread function. The axioms extended from the 1-D counterparts include nonnegativity, continuity, translation invariance, rotation invariance, luminance invariance, homogeneous scaling, and serial combination properties. It is proved that the only resolution measure consistent with the axioms is proportional to the square root of the trace of the covariance matrix of the point spread function.

This editorial introduces Book Reviews as a new publication category for the IEEE Transactions on Medical Imaging (T-MI). Recent progress in establishment of the National Institutes of Health (NIH) National Institute of Biomedical Imaging and Bioengineering is summarized. An Institute of Medicine report on emerging technologies for breast cancer detection and screening is announced.

Three-dimensional (3-D) modeling and visualization of the cochlea using the World Wide Web (WWW) is an effective way of sharing anatomic information for cochlear implantation over the Internet, particularly for morphometry-based research and resident training in otolaryngology and neuroradiology. In this paper, 3-D modeling, visualization, and animation techniques are integrated in an interactive and platform-independent manner and implemented over the WWW. Cohen's template shape with mean cross-sectional areas of the human cochlea is extended into a 3-D geometrical model. Also, spiral computer tomography data of a patient's cochlea is digitally segmented and geometrically represented. The cochlear electrode array is synthesized according to its specification. Then, cochlear implantation is animated with both idealized and real cochlear models. Insertion length, angular position, and characteristic frequency of individual electrodes are estimated online during the virtual insertion. The optimization of the processing parameters is done to demonstrate the feasibility of this technology for clinical applications.

Traditional computed tomography (CT) reconstructions of total joint prostheses are limited by metal artifacts from corrupted projection data. Published metal artifact reduction methods are based on the assumption that severe attenuation of X-rays by prostheses renders corresponding portions of projection data unavailable, hence the "missing" data are either avoided (in iterative reconstruction) or interpolated (in filtered backprojection with data completion; typically, with filling data "gaps" via linear functions). In this paper, we propose a wavelet-based multiresolution analysis method for metal artifact reduction, in which information is extracted from corrupted projection data. The wavelet method improves image quality by a successive interpolation in the wavelet domain. Theoretical analysis and experimental results demonstrate that the metal artifacts due to both photon starving and beam hardening can be effectively suppressed using our method. As compared to the filtered backprojection after linear interpolation, the wavelet-based reconstruction is significantly more accurate for depiction of anatomical structures, especially in the immediate neighborhood of the prostheses. This superior imaging precision is highly advantageous in geometric modeling for fitting hip prostheses.

X-ray computed tomography (CT) is in transition from fan-beam to cone-beam geometry. For cone-beam volumetric imaging, reduction of radiation exposure remains an important issue. Because the wavelet approach was shown to be effective and flexible for two-dimensional (2-D) local region reconstruction, we are motivated to perform wavelet local CT in cone-beam geometry. In this paper, we formulate the Feldkamp cone-beam reconstruction from the wavelet perspective, derive both full-scan and half-scan Feldkamp-type formulas for either global or local reconstruction, and demonstrate the feasibility and utility in synthetic and real data. It is found that using the wavelet Feldkamp approach, a three-dimensional (3-D) region of interest (ROI) can be reconstructed with neither severe image artifacts nor any significant constant bias in our simulation and experiments.

In signal processing, reconstruction of a bandlimited signal from finite samples is complicated by the underdetermined nature of the problem and the unavoidable noise in the measurement. A sensitivity function was recently formulated assuming noise-free data, which provides point-wise information about reliability of the reconstructed signal before actual samples of the signal are taken. In this communication, the minimum error bound of signal reconstruction is derived assuming noisy data.

Image resolution is the primary parameter for performance characterization of any imaging system. In this work, we present an axiomatic approach for quantification of image resolution, and demonstrate that a good image resolution measure should be proportional to the standard deviation of the point spread function of an imaging system.

Gastrointestinal (GI) tract examination with spiral/helical computed tomography (CT) is currently performed by slice-based inspection of axial images. CT colography is a recent advance which allows an intraluminal visualization of the colon, similar to endoscopy. Various rendering algorithms have been developed with promising results, however navigation through the complex, tortuous anatomy of the colon can be time consuming in practice. In this paper, we propose an electrical-field-based method to unravel the convoluted colon, that is, to digitally straighten it with curved cross sections and flatten it over a plane. In our method, electrical charges are simulated along the central colon path. Curved cross sections are defined by the electrical force lines, and lead to consistent unraveling. It is demonstrated with image volumes of two patients that this technique produces a global planar view of complicated colon features with a potential for detection of polyps.

X-ray computed tomography fluoroscopy (CTF) enables image guidance of interventions, synchronization of scanning with contrast bolus arrival, and motion analysis. However, filtered backprojection (FB), the current method for CTF image reconstruction, is subject to motion and metal artifacts from implants, needles, or other surgical instruments. Reduced target lesion conspicuity may result from increased image noise associated with reduced tube current. In this report, we adapt the row-action expectation-maximization (EM) algorithm for CTF. Because time-dependent variation in images is localized during CTF, the row-action EM-like algorithm allows rapid convergence. More importantly, this iterative CTF algorithm has fewer metal artifacts and better low-contrast performance than FB.

Cochlear implantation is the standard treatment for profound hearing loss. Preimplantation and postimplantation spiral computed tomography (CT) is essential in several key clinical and research aspects. The maximum image resolution with commercial spiral CT scanners is insufficient to define clearly anatomical features and implant electrode positions in the inner ear. In this paper, we develop an expectation-maximization (EM)-like iterative deblurring algorithm to achieve spiral CT image super-resolution for cochlear implantation, assuming a spatially invariant linear spiral CT system with a three-dimensional (3-D) separable Gaussian point spread function (PSF). We experimentally validate the 3-D Gaussian blurring model via phantom measurement and profile fitting. The imaging process is further expressed as convolution of an isotropic 3-D Gaussian PSF and a blurred underlying volumetric image. Under practical conditions, an oblique reconstructed section is approximated as convolution of an isotropic two-dimensional (2-D) Gaussian PSF and the corresponding actual cross section. The spiral CT image deblurring algorithm is formulated with sieve and resolution kernels for suppressing noise and edge artifacts. A typical cochlear cross section is used for evaluation, demonstrating a resolution gain up to 30%40% according to the correlation criterion. Physical phantoms, preimplantation and postimplantation patients are reconstructed into volumes of 0.1-mm cubic voxels. The patient images are digitally unwrapped along the central axis of the cochlea and the implanted electrode array respectively, then oblique sections orthogonal to the central axis formed. After deblurring, representation of structural features is substantially improved in all the cases.

Iterative deblurring methods using the expectation maximization (EM) formulation and the algebraic reconstruction technique (ART), respectively, are adapted for metal artifact reduction in medical computed tomography (CT). In experiments with synthetic noise-free and additive noisy projection data of dental phantoms, it is found that both simultaneous iterative algorithms produce superior image quality as compared to filtered backprojection after linearly fitting projection gaps. Furthermore, the EM-type algorithm converges faster than the ART-type algorithm in terms of either the I-divergence or Euclidean distance between ideal and reprojected data in the authors' simulation. Also, for a given iteration number, the EM-type deblurring method produces better image clarity but stronger noise than the ART-type reconstruction. The computational complexity of EM- and ART-based iterative deblurring is essentially the same, dominated by reprojection and backprojection. Relevant practical and theoretical issues are discussed.

Helical computed tomography (helical/spiral CT) has replaced conventional CT in many clinical applications. In current helical CT, a patient is assumed to be rigid and motionless during scanning and planar projection sets are produced from raw data via longitudinal interpolation. However, rigid patient: motion is a problem in some cases (such as in the skull base and temporal bone imaging). Motion artifacts thus generated in reconstructed images can prevent accurate diagnosis. Modeling a uniform translational movement, the authors address how patient motion is ascertained and how it may be compensated. First, mismatch between adjacent fan-beam projections of the same orientation is determined via classical correlation, which is approximately proportional to the patient displacement projected onto an axis orthogonal to the central ray of the involved fan-beam. Then, the patient motion vector (the patient displacement per gantry rotation) is estimated from its projections using a least-square-root method. To suppress motion artifacts, adaptive interpolation algorithms are developed that synthesize full-scan and half-scan planar projection data sets, respectively. In the adaptive scheme, the interpolation is performed along inclined paths dependent upon the patient motion vector. The simulation results show that the patient motion vector can be accurately and reliably estimated using the authors' correlation and least-square-root algorithm, patient motion artifacts can be effectively suppressed via adaptive interpolation, and adaptive half-scan interpolation is advantageous compared with its full-scan counterpart in terms of high contrast image resolution.

In order to perform fan-beam reconstruction using projection data collected from a noncircular scanning locus, existing noncircular fan-beam formulas require a derivative of the scanning locus with respect to the rotation angle. A derivative-free noncircular fan-beam reconstruction formula that is based on a geometrical explanation of the circular equispatial fan-beam reconstruction formula is obtained here. A mathematical proof is then provided under the conditions that the source-to-origin distance is symmetric with respect to the origin of the reconstruction coordinate system, is differentiable almost everywhere, and does not change too fast with respect to the rotation angle. The derivative-free noncircular fan-beam reconstruction formula is the same as the circular one, except that the source-to-origin distance is a function of the rotation angle. A typical simulation result for the noncircular fan-beam formula is given.

Considering the characteristics of the X-ray microscope system being developed at SUNY at Buffalo and the limitations of available cone-beam reconstruction algorithms, a general cone-beam reconstruction algorithm and several special versions of it are proposed and validated by simulation. The cone-beam algorithm allows various scanning loci, handles reconstruction of rod-shaped specimens which are common in practice, and facilitates near real-time reconstruction by providing the same computational efficiency and parallelism as L.A. Feldkamp et al.'s (1984) algorithm. Although the present cone-beam algorithm is not exact, it consistently gives satisfactory reconstructed images. Furthermore, it has several nice properties if the scanning locus meets some conditions. First, reconstruction within a midplane is exact using a planar scanning locus. Second, the vertical integral of a reconstructed image is equal to that of the actual image. Third, reconstruction is exact if an actual image is independent of rotation axis coordinate z. Also, the general algorithm can uniformize and reduce z-axis artifacts, if a helix-like scanning locus is used.