The segmentation of 2D vascular structures via deep learning holds significant clinical value but is hindered by the scarcity of annotated data, severely limiting its widespread application. Developing a universal few-shot vascular segmentation model is highly desirable, yet remains challenging due to the need for extensive training and the inherent complexities of vascular imaging. In this work, we propose UniVG (Generative Data-engine Foundation Model for Universal Few-shot 2D Vascular Image Segmentation), a novel approach that learns the compositionality of vascular images and constructing a generative foundation model for robust vascular segmentation. UniVG enables the synthesis and learning of diverse and realistic vascular images through two key innovations: 1) Compositional learning for flexible and diverse vascular synthesis: It decomposes and recombines vascular structures with varying morphological features and diverse foreground-background configurations to generate richly diverse synthetic image-label pairs. 2) Few-shot generative adaptation for transferable segmentation: It fine-tunes pre-trained models with minimal annotated data to bridge the gap between synthetic and real vascular domains, synthesizing authentic and diverse vessel images for downstream few-shot vascular segmentation learning. To support our approach, we develop UniVG-58K, a large dataset comprising 58,689 vascular images across five imaging modalities, facilitating robust large-scale generative pre-training. Extensive experiments on 11 vessel segmentation tasks cross 5 modalties (only with 5 labeled images on each task) demonstrate that UniVG achieves performance comparable to fully supervised models, significantly reducing data collection and annotation costs. All code and datasets will be made publicly available at https://github.com/XinAloha/UniVG.

Segment Anything Model (SAM) has demonstrated state-of-the-art performance in most segmentation tasks. However, due to insufficient training in the medical domain, SAM’s ability to generalize to medical images is limited. Although preliminary efforts have fine-tuned SAM for the medical domain, the fine-tuned model still struggles with variability in medical tasks. Some recent studies have explored weakly supervised learning to mitigate SAM’s performance degradation in the medical domain. However, the effectiveness of weakly supervised learning is heavily dependent on the quality of weakly supervised information, with performance significantly dropping as the quality declines. Doctors’ attention is closely related to the target area during diagnosis. Integrating gaze information into SAM’s adaptation process for medical image segmentation enhances efficiency and significantly improves performance in medical tasks. In this paper, we first propose a Gaze-assisted medical segment Anything Model (GAM), which utilizes gaze information to enable the adaptation of SAM in medical images following doctor’s attention. It has two innovations: 1) Feature-level adaptation: Gaze Alignment (GA) learning makes the feature-level adaptation follow the doctor’s attention which mines the human guidance from gaze heatmaps and guides model to extract general features for downstream tasks. 2) Output-level adaptation: Gaze-Balance (GB) learning makes the output-level adaptation follow the doctor’s attention which utilizes gaze heatmaps to enhance the human-focused area and solve the problem of over/under segmentation from the output-level. Our promising results on 7 tasks with 12 targets have demonstrated the powerful adaptation ability of our GAM in the medical domain. Our GAM demonstrates significant potential for low-cost clinical assistance in medical diagnosis, enabling SAM to adapt to the medical image domain without disrupting clinical workflows. We have released the full source code on https://github.com/Ruiz1026/GAM.

In the field of medical image processing, vascular image segmentation plays a crucial role in clinical diagnosis, treatment planning, prognosis, and medical decision-making. Accurate and automated segmentation of vascular images can assist clinicians in understanding the vascular network structure, leading to more informed medical decisions. However, manual annotation of vascular images is time-consuming and challenging due to the fine and low-contrast vascular branches, especially in the medical imaging domain where annotation requires specialized knowledge and clinical expertise. Data-driven deep learning models struggle to achieve good performance when only a small number of annotated vascular images are available. To address this issue, this paper proposes a novel Conditional Virtual Imaging (CVI) framework for few-shot vascular image segmentation learning. The framework combines limited annotated data with extensive unlabeled data to generate high-quality images, effectively improving the accuracy and robustness of segmentation learning. Our approach primarily includes two innovations: First, aligned image-mask pair generation, which leverages the powerful image generation capabilities of large pre-trained models to produce high-quality vascular images with complex structures using only a few training images; Second, the Dual-Consistency Learning (DCL) strategy, which simultaneously trains the generator and segmentation model, allowing them to learn from each other and maximize the utilization of limited data. Experimental results demonstrate that our CVI framework can generate high-quality medical images and effectively enhance the performance of segmentation models in few-shot scenarios. Our code will be made publicly available online.

Ultrasound image segmentation is critical for tumor screening but is hindered by noise, artifacts, and high variability in lesion appearance. Challenges like blurred boundaries and morphological similarities further complicate accurate delineation. To address this, we propose the Dynamic Edge-aware Transformer with Cross-image Patch Dependency (DET-CPD). Our model integrates two key modules: a Dynamic Difference Convolution Module (DDCM) to enhance edge representation for varied lesions, and a Cross-Scale Semantic Enhancement Module (CSEM) that leverages cross-scale channel information to distinguish tumors from surrounding tissue. Crucially, we introduce a novel Cross-image Patch Dependency Loss (CPDLoss) that captures semantic dependencies across different images in a batch, improving robustness. Extensive experiments on four public datasets (BUSI, DatasetB, DDTI, and TN3K) demonstrate that DET-CPD achieves state-of-the-art segmentation performance.

Accurate automated measurement of diaphragmatic thickness in ultrasound imaging is a critical challenging task for respiratory function assessment, primarily due to difficulties in precise fascial identification. And ultrasound visualization of the diaphragm is characterized by unique challenges, including discontinuous and blurred boundary delineations caused by imaging artifacts, as well as interference and influence from adjacent muscular reverberations. These problems are further compounded by subjects’ pose variations during image acquisition. To address these challenges, we introduce IBS-Net, an innovative triple-branch interactive segmentation network that synergistically combines boundary regression with auxiliary task learning to optimize feature representation in segmentation task. Moreover, Our framework incorporates two innovative module: an Adaptive Fusion Module (AFM) that enables multi-scale hierarchical feature refinement for precise boundary characterization, and a Cross Interactive Module (CIM) that employs parallel-encoded feature extraction to simultaneously achieve accurate fascial localization while preserving structural topology. These complementary mechanisms effectively resolve spatial feature inconsistencies, facilitating robust multi-level feature integration. Comprehensive experimental results demonstrate that IBS-Net achieves statistically significant improvements of 8.9% in Dice similarity coefficient and 8.05% in Jaccard index compared to conventional methods. Moreover, to verify the effectiveness of the proposed method, we extended it to other publicly available BUSI dataset for experimentation. The results demonstrate that our method is competitive in terms of both accuracy and completeness in the identification of fuzzy boundaries in ultrasound images.

Domain adaptation (DA) for cardiac ultrasound image segmentation is clinically significant and valuable. However, previous domain adaptation methods are prone to be affected by the incomplete pseudo label and low-quality target to source images. Human-centric domain adaptation has great advantages of human cognitive guidance to help model adapt to target domain and reduce reliance on labels. Doctor gaze trajectories contains a large amount of cross-domain human guidance. To leverage gaze information and human cognition for guiding domain adaptation, we propose gaze-assisted human-centric domain adaptation (GAHCDA), which reliably guides the domain adaptation of cardiac ultrasound images. GAHCDA includes following modules: (1) Gaze Augment Alignment (GAA): GAA enables the model to obtain human cognition general features to recognize segmentation target in different domain of cardiac ultrasound images like humans. (2) Gaze Balance Loss (GBL): GBL fused gaze heatmap with outputs which makes the segmentation result structurally closer to the target domain. The experimental results show that our proposed framework is able to segment cardiac ultrasound images more effectively in the target domain than GAN-based methods and other self-train based methods and shown great potential in clinical application.

Dense contrastive representation learning (DCRL) has greatly improved the learning efficiency for image dense prediction tasks, showing its great potential to reduce the large costs of medical image collection and dense annotation. However, the properties of medical images make unreliable correspondence discovery, bringing an open problem of large-scale false positive and negative (FP&N) pairs in DCRL. In this paper, we propose GEoMetric vIsual deNse sImilarity (GEMINI) learning which embeds the homeomorphism prior to DCRL and enables a reliable correspondence discovery for effective dense contrast. We proposes a deformable homeomorphism learning (DHL) which models the homeomorphism of medical images and learns to estimate a deformable mapping to predict the pixels' correspondence under the condition of topological preservation. It effectively reduces the searching space of pairing and drives an implicit and soft learning of negative pairs via gradient. We also proposes a geometric semantic similarity (GSS) which extracts semantic information in features to measure the alignment degree for the correspondence learning. It will promote the learning efficiency and performance of deformation, constructing positive pairs reliably. We implement two practical variants on two typical representation learning tasks in our experiments. Our promising results on seven datasets which outperform the existing methods show our great superiority. We will release our code at a companion website.

In this work, we address the task of few-shot medical image segmentation (MIS) with a novel proposed framework based on the learning registration to learn segmentation (LRLS) paradigm. To cope with the limitations of lack of authenticity, diversity, and robustness in the existing LRLS frameworks, we propose the better registration better segmentation (BRBS) framework with three main contributions that are experimentally shown to have substantial practical merit. First, we improve the authenticity in the registration-based generation program and propose the knowledge consistency constraint strategy that constrains the registration network to learn according to the domain knowledge. It brings the semantic-aligned and topology-preserved registration, thus allowing the generation program to output new data with great space and style authenticity. Second, we deeply studied the diversity of the generation process and propose the space-style sampling program, which introduces the modeling of the transformation path of style and space change between few atlases and numerous unlabeled images into the generation program. Therefore, the sampling on the transformation paths provides much more diverse space and style features to the generated data effectively improving the diversity. Third, we first highlight the robustness in the learning of segmentation in the LRLS paradigm and propose the mix misalignment regularization, which simulates the misalignment distortion and constrains the network to reduce the fitting degree of misaligned regions. Therefore, it builds regularization for these regions improving the robustness of segmentation learning. Without any bells and whistles, our approach achieves a new state-of-the-art performance in few-shot MIS on two challenging tasks that outperform the existing LRLS-based few-shot methods. We believe that this novel and effective framework will provide a powerful few-shot benchmark for the field of medical image and efficiently reduce the costs of medical image research. All of our code will be made publicly available online.

Learning inter-image similarity is crucial for 3D medical images self-supervised pre-training, due to their sharing of numerous same semantic regions. However, the lack of the semantic prior in metrics and the semantic-independent variation in 3D medical images make it challenging to get a reliable measurement for the inter-image similarity, hindering the learning of consistent representation for same semantics. We investigate the challenging problem of this task, i.e., learning a consistent representation between images for a clustering effect of same semantic features. We propose a novel visual similarity learning paradigm, Geometric Visual Similarity Learning, which embeds the prior of topological invariance into the measurement of the inter-image similarity for consistent representation of semantic regions. To drive this paradigm, we further construct a novel geometric matching head, the Z-matching head, to collaboratively learn the global and local similarity of semantic regions, guiding the efficient representation learning for different scale-level inter-image semantic features. Our experiments demonstrate that the pre-training with our learning of inter-image similarity yields more powerful inner-scene, inter-scene, and global-local transferring ability on four challenging 3D medical image tasks. Our codes and pre-trained models will be publicly available11https://github.com/YutingHe-list/GVSL.

Low-Dose Computed Tomography (LDCT) technique, which reduces the radiation harm to human bodies, is now attracting increasing interest in the medical imaging field. As the image quality is degraded by low dose radiation, LDCT exams require specialized reconstruction methods or denoising algorithms. However, most of the recent effective methods overlook the inner-structure of the original projection data (sinogram) which limits their denoising ability. The inner-structure of the sinogram represents special characteristics of the data in the sinogram domain. By maintaining this structure while denoising, the noise can be obviously restrained. Therefore, we propose an LDCT denoising network namely Sinogram Inner-Structure Transformer (SIST) to reduce the noise by utilizing the inner-structure in the sinogram domain. Specifically, we study the CT imaging mechanism and statistical characteristics of sinogram to design the sinogram inner-structure loss including the global and local inner-structure for restoring high-quality CT images. Besides, we propose a sinogram transformer module to better extract sinogram features. The transformer architecture using a self-attention mechanism can exploit interrelations between projections of different view angles, which achieves an outstanding performance in sinogram denoising. Furthermore, in order to improve the performance in the image domain, we propose the image reconstruction module to complementarily denoise both in the sinogram and image domain.

Sparse-View CT (SVCT), which provides low-dose and high-speed CT imaging, plays an important role in the medical imaging area. As the decrease of projection views, the reconstructed image suffers from severe artifacts. To this end, recent works utilize deep learning methods to improve the imaging quality of SVCT and achieve promising performances. However, these methods mainly focus on the network design and modeling but overlook the importance of choosing projection views. To address this issue, this paper proposes a Projection-view LeArning Network (PLANet), which can estimate the importance of different view angles through reconstruction network training and select the projection views for high-quality image restoration. Specifically, we generate synthesized sparse-view sinograms by subsampling projections from full-view sinograms based on a learnable distribution, which can be learned through reconstruction network training. Thus, important image views can be selected to acquire sparse-view projection in imaging equipment. Furthermore, effective data augmentations are provided by the online generation of sparse-view sinogram to improve the stability and performance of reconstruction networks. In short, our method can select the important projection views and learn high-performance reconstruction networks in one unified deep-learning framework. Comprehensive experiments show that the proposed method achieves promising results compared to state-of-the-art methods, and the ablation studies also show the superiority of our proposed PLANet in terms of effectiveness and robustness.

Deformable medical image registration estimates corresponding deformation to align the regions of interest (ROIs) of two images to a same spatial coordinate system. However, recent unsupervised registration models only have correspondence ability without perception, making misalignment on blurred anatomies and distortion on task-unconcerned backgrounds. Label-constrained (LC) registration models embed the perception ability via labels, but the lack of texture constraints in labels and the expensive labeling costs causes distortion internal ROIs and overfitted perception. We propose the first few-shot deformable medical image registration framework, Perception-Correspondence Registration (PC-Reg), which embeds perception ability to registration models only with few labels, thus greatly improving registration accuracy and reducing distortion. 1) We propose the Perception-Correspondence Decoupling which decouples the perception and correspondence actions of registration to two CNNs. Therefore, independent optimizations and feature representations are available avoiding interference of the correspondence due to the lack of texture constraints. 2) For few-shot learning, we propose Reverse Teaching which aligns labeled and unlabeled images to each other to provide supervision information to the structure and style knowledge in unlabeled images, thus generating additional training data. Therefore, these data will reversely teach our perception CNN more style and structure knowledge, improving its generalization ability. Our experiments on three datasets with only five labels demonstrate that our PC-Reg has competitive registration accuracy and effective distortion-reducing ability. Compared with LC-VoxelMorph( λ = 1), we achieve the 12.5%, 6.3% and 1.0% Reg-DSC improvements on three datasets, revealing our framework with great potential in clinical application.

Susceptibility weighted imaging (SWI) is a routine magnetic resonance imaging (MRI) sequence that combines the magnitude and high-pass filtered phase images to qualitatively enhance the image contrasts related to tissue susceptibility. Tremendous amounts of the high-pass filtered phase data with low signal to noise ratio and incomplete background field removal have thus been collected under default clinical settings. Since SWI cannot quantitatively estimate the susceptibility, it is thus non-trivial to derive quantitative susceptibility mapping (QSM) directly from these redundant phase data, which effectively promotes the mining of the SWI data collected previously. To this end, a novel deep learning based SWI-to-QSM-Net (S2Q-Net) is proposed for QSM reconstruction from SWI high-pass filtered phase data. S2Q-Net firstly estimates the edge maps of QSM to integrate edge prior into features, which benefits the network to reconstruct QSM with realistic and clear tissue boundaries. Furthermore, a novel Second-order Cross Dense Block is proposed in S2Q-Net, which can capture rich inter-region interactions to provide more non-local phase information related to local tissue susceptibility. Experimental results on both simulated and in-vivo data indicate its superiority over all the compared deep learning based QSM reconstruction methods.

Receptive field-based segmentation models represent features in receptive fields having weak perception for thin semantics in thin structures segmentation, due to the challenges in small local size and large global variation. High-frequency (HiFe) components have strong thin perception ability and is stable for global variation, but its weak adaptability limits its direct application. We propose a HiFe priori rule which enables the network to adaptively extract and fuse HiFe components, enhancing the thin semantics and making the network naturally prefer thin structures for their segmentation. We further propose High-Frequency Semantics Enhancement Network (HiFeNet) based on our HiFe priori rule, boosting the SOTA methods in thin structures segmentation: 1) Our Deep High Frequency (DHiFe) block learns to extract task-dependent HiFe components and adds them to feature maps, achieving great perception of thin structures. 2) Our Latent Residual Denoising (LRD) block progressively weakens task-independent features via hierarchical residuals and learns to fuse HiFe components back to feature maps, further enhancing the thin semantics and weakening the interference of global variation. Extensive experiments on the retinal vessel [1], [2], [3] and Massachusetts road [4] segmentation datasets show great superiority of our HiFeNet.

The integration of segmentation and direct quantification on the left ventricle (LV) from the paired apical views(i.e., apical 4-chamber and 2-chamber together) echo sequence clinically achieves the comprehensive cardiac assessment: multiview segmentation for anatomical morphology, and multidimensional quantification for contractile function. Direct quantification of LV, i.e., to automatically quantify multiple LV indices directly from the image via task-aware feature representation and regression, avoids accumulative error from the inter-step target. This integration sequentially makes a stereoscopical reflection of cardiac activity jointly from the paired orthogonal cross views sequences, overcoming limited observation with a single plane. We propose a K-shaped Unified Network (K-Net), the first end-to-end framework to simultaneously segment LV from apical 4-chamber and 2-chamber views, and directly quantify LV from major- and minor-axis dimensions (1D), area (2D), and volume (3D), in sequence. It works via four components: 1) the K-Net architecture with the Attention Junction enables heterogeneous tasks learning of segmentation task of pixel-wise classification, and direct quantification task of image-wise regression, by interactively introducing the information from segmentation to jointly promote spatial attention map to guide quantification focusing on LV-related region, and transferring quantification feedback to make global constraint on segmentation; 2) the Bi-ResLSTMs distributed in K-Net layer-by-layer hierarchically extract spatial-temporal information in echo sequence, with bidirectional recurrent and short-cut connection to model spatial-temporal information among all frames; 3) the Information Valve tailing the Bi-ResLSTMs selectively exchanges information among multiple views, by stimulating complementary information and suppressing redundant information to make the efficient cross-flow for each view; 4) the Evolution Loss comprehensively guides sequential data learning, with static constraint for frame values, and dynamic constraint for inter-frame value changes. The experiments show that our K-Net gains high performance with a Dice coefficient up to 91.44% and a mean absolute error of the major-axis dimension down to 2.74mm, which reveal its clinical potential.