Yuanjing Feng

Accurate delineation of cerebrovascular structures from Time‐Of‐Flight Magnetic Resonance Angiography (TOF‐MRA) and Computed Tomography Angiography (CTA) is essential for the clinical diagnosis and treatment of cerebrovascular diseases. However, the intricate topology and fine‐scale nature of cerebral vessels pose significant challenges for deep learning methods, which often struggle to capture long‐range dependencies and precise morphological details. In this work, we propose VesMamba, a deep learning framework that integrates explicit vascular morphological priors into a state‐space model. Unlike generic SSM‐based methods that rely on fixed scanning patterns, we introduce a Tri‐oriented Vessel‐aware Mamba (ToVM) module, which dynamically reorders input 1D sequences using cerebrovascular edge features to better model complex vascular structures. Complementarily, we present the 3D Large‐Small Gated Convolution (LSGC) module after the ToVM module to preserve critical spatial information. We conducted extensive experiments on two TOF‐MRA and one CTA dataset, comparing our method with eight state‐of‐the‐art approaches. Our results show that VesMamba achieves superior performance on the majority of evaluation metrics relative to all competing methods.

Compressive sensing (CS) has revolutionized image acquisition and reconstruction; however, its primary application has been in the domain of single-sence image processing. This study introduces a novel approach that extends CS frameworks to multi-scene settings, enabling the efficient separation and reconstruction of multiple distinct scenes from fused random measurements. The key innovation lies in the theoretical derivation of the momentum weight structure for cross-scene fusion, which is proven to be the sum of divergences from two proximal mapping processes. The proposed method deconstructs the multi-scene CS challenge into multiple sub-optimization tasks with interconnected message passing, where each scene’s reconstruction is solved using an iterative alternating-minimization / block-coordinate descent scheme. Our multi-scene approach enables multiple sensors to share a single transmission channel, balancing the reconstruction quality across multiple scenes. The experimental results demonstrate that our multi-scene compressed sensing framework, integrated with an efficient deep image denoiser as the proximal operator, achieves superior reconstruction quality. The test code is available at https://github.com/ZanChen1/Multi-scene-CS.

Low-dose X-ray fluoroscopy is essential in interventional procedures to reduce radiation exposure, yet image quality is inevitably degraded by predominantly Poisson noise, compromising diagnostic reliability. Existing denoising methods commonly rely on large annotated medical datasets, which are scarce in practice, and struggle to balance noise suppression with the preservation of fine structural details. To overcome the above-mentioned problems, we propose SIX, a zero-shot self-supervised interactive denoising framework for low-dose fluoroscopy. The framework integrates a physics-driven noise model with a deep image prior. It performs prior-based denoising via probabilistic modeling with an intuitive hyperparameter, allowing a flexible balance between noise suppression and structural fidelity according to specific diagnostic requirements. Experiments on an in-house dataset show that SIX outperforms various conventional and deep learning methods in PSNR and SSIM, as well as in clinically critical metrics including contrast-to-noise ratio and spatial resolution. With its flexibility and user-friendly design, SIX holds strong potential for deployment in clinical image-guided workflows.

Positron emission tomography (PET) image enhancement is important for improving clinical diagnostic accuracy. However, most existing enhancement methods rely on paired datasets, which are difficult to acquire in clinical practice. This study aims to develop a robust unpaired framework for PET image enhancement that can effectively leverage the widely available unpaired low- and high-quality datasets. By eliminating the need for paired data, the proposed framework can generalize across diverse scanners, protocols, and radiotracer doses. We propose a two-stage diffusion-based framework for unpaired PET image enhancement. In Stage 1, an unconditional diffusion model is trained on low-quality data to learn their distribution. During sampling, unpaired high-quality data guides the generation of pseudo low-quality counterparts via a conditional strategy combining timestep truncation and frequency-domain constraints, constructing a synthetic paired dataset. In Stage 2, a denoising network (e.g., RED-CNN) is trained on this synthetic dataset for efficient enhancement. The proposed method consistently outperforms traditional approaches (BM3D, BM4D), unsupervised diffusion-based methods (ILVR, SDEdit, EGSDE), self-supervised denoising methods (Noise2Void, Conditional Deep Image Prior(CDIP)), and unpaired methods (IE-CycleGAN) on both low-quality brain datasets from the uMI 780 scanner (comprising 420 subjects) and the Biograph mMR scanner (comprising 19 subjects). Compared with supervised methods, our approach demonstrates superior generalization ability and exhibits better applicability to real-world clinical datasets lacking paired training data. The proposed method effectively harnesses the powerful generative capacity of diffusion models while bypassing the time-consuming inference issue inherent to direct application of diffusion models through a flexible second stage. This provides a robust and scalable framework for enhancing real-world low-quality PET images without requiring paired data.

Diffusion magnetic resonance imaging (dMRI) is a non-invasive technique used to characterize tissue microstructure by measuring the diffusion of water molecules. Conventional Q-space trajectory imaging (QTI) estimates diffusion using low-order moments; however, it often neglects higher-order moments, such as the skewness tensor, resulting in an incomplete representation of diffusion asymmetry and potential estimation bias. In this work, we propose Q-space trajectory imaging with Skewness Tensor Constraints (QTI-STC), a method that incorporates higher-order skewness tensors under positivity constraints to mitigate deviations in the estimation of lower-order moments caused by the omission of higher-order asymmetry information. Furthermore, we introduce linear trace-weighted (LF) and quadratic trace-weighted filters (QF) to enhance high-diffusion components while suppressing low-diffusion components. Extensive experiments conducted on public, noisy, and synthetic datasets demonstrate that our method yields estimates closer to the ground truth on synthetic data and exhibits superior robustness in noisy conditions.

Objective: Cranial nerves (CNs) play a crucial role in various essential functions of the human brain, and mapping their pathways from diffusion MRI (dMRI) provides valuable preoperative insights into the spatial relationships between individual CNs and key tissues. However, mapping a comprehensive and detailed CN atlas is challenging because of the unique anatomical structures of each CN pair and the complexity of the skull base environment. Method: In this work, we present what we believe to be the first study to develop a comprehensive diffusion tractography atlas for automated mapping of CN pathways in the human brain. The CN atlas is generated by fiber clustering by using the streamlines generated by multi-parametric fiber tractography for each pair of CNs. Instead of disposable clustering, we explore a new strategy of multi-stage fiber clustering for multiple analysis of approximately 1,000,000 streamlines generated from the 50 subjects from the Human Connectome Project (HCP). Results: Quantitative and visual experiments demonstrate that our CN atlas achieves high spatial correspondence with expert manual annotations on multiple acquisition sites, including the HCP dataset, the Multi-shell Diffusion MRI (MDM) dataset and two clinical cases of pituitary adenoma patients. Conclusion: The proposed CN atlas can automatically identify 8 fiber bundles associated with 5 pairs of CNs, including the optic nerve CN II, oculomotor nerve CN III, trigeminal nerve CN V and facial-vestibulocochlear nerve CN VII/VIII, and its robustness is demonstrated experimentally. Significance: This work contributes to the field of diffusion imaging by facilitating more efficient and automated mapping the pathways of multiple pairs of CNs, thereby enhancing the analysis and understanding of complex brain structures through visualization of their spatial relationships with nearby anatomy.

Existing compressive sensing (CS) reconstruction algorithms are primarily designed for deterministic measurements. However, in real-world scenarios, quantization bit erasures during transmission or storage introduce uncertainty into measurements and significantly complicate reconstruction. We extend the CS framework to explicitly handle such bit-level uncertainty, enabling robust image recovery from quantized measurements with missing bits. We begin by enumerating all valid combinations of the erased quantization bits to construct a candidate set of feasible measurement values. This candidate set is then incorporated as a constraint in a newly formulated inverse problem. We propose an iterative plug-and-play algorithm to solve this problem, alternating between two key steps: (1) an image update using a pretrained denoiser, and (2) a measurement update via a soft-min projection strategy accelerated by coordinate descent. Extensive experiments demonstrate the effectiveness of the proposed approach, achieving high-quality reconstructions even under extremely low sampling rates and severe erasure conditions. Our framework offers a scalable and principled solution for bit-erasure-robust CS reconstruction in error-prone and resource-constrained imaging environments. The test code is available at https://github.com/ZanChen1/RCS-QuantBitErasure.

Objective. Accurately delineating the visual pathway (VP) is crucial for understanding the human visual system and diagnosing related disorders. Exploring multi-parametric MR imaging data has been identified as an important way to delineate VP. However, due to the complex cross-sequence relationships, existing methods cannot effectively model the complementary information from different MRI sequences. In addition, these existing methods heavily rely on large training data with labels, which is labor-intensive and time-consuming to obtain. Approach. We propose a novel semi-supervised multi-parametric feature decomposition framework for VP delineation. Specifically, a correlation-constrained feature decomposition is designed to handle the complex cross-sequence relationships by capturing the unique characteristics of each MRI sequence and easing the multi-parametric information fusion process. Furthermore, a consistency-based sample enhancement module is developed to address the limited labeled data issue, by generating and promoting meaningful edge information from unlabeled data. Main results. We validate our framework using two public datasets and one in-house multi-shell diffusion MRI dataset. Experimental results demonstrate the superiority of our approach in terms of delineation performance when compared to six state-of-the-art approaches. Significance. Our proposed framework effectively mitigates the challenges of modeling complex cross-sequence relationships and limited labeled data, offering a robust solution for accurate VP delineation. This approach not only enhances the understanding of the human visual system but also holds potential for improving the diagnosis of VP-related disorders.