Testis size is known to be one of the main predictors of male fertility, usually assessed in clinical workup via palpation or imaging. Despite its potential, population-level evaluation of testicular volume using imaging remains underexplored. Previous studies, limited by small and biased datasets, have demonstrated the feasibility of machine learning for testis volume segmentation. This paper presents an evaluation of segmentation methods for testicular volume using Magnetic Resonance Imaging data from the UKBiobank. The best model achieves a median dice score of 0.89, compared to median dice score of 0.85 for human interrater reliability on the same dataset, enabling large-scale annotation on a population scale for the first time. Our overall aim is to provide a trained model, comparative baseline methods, and annotated training data to enhance accessibility and reproducibility in testis MRI segmentation research.

Medical image segmentation is a critical task for the early detection and diagnosis of various conditions, such as skin cancer, polyps, thyroid nodules, and pancreatic tumors. Recently, deep learning architectures have achieved significant success in this field. However, they face a critical trade-off between local feature extraction and global context modeling. To address this limitation, we present DCM-Net, a dual-encoder architecture that integrates pretrained CNN layers with Visual State Space (VSS) blocks through a Cross-Branch Feature Fusion Module (CBFFM). A Decoder Feature Enhancement Module (DFEM) combines depth-wise separable convolutions with MLP-based semantic rectification to extract enhanced decoded features and improve the segmentation performance. Additionally, we present a new 2D pancreas and pancreatic tumor dataset (CCH-PCT-CT) collected from Chongqing University Cancer Hospital, comprising 3,547 annotated CT slices, which is used to validate the proposed model. The proposed DCM-Net architecture achieves competitive performance across all datasets investigated in this study. We develop a novel DCM-Net architecture that generates robust features for tumor and organ segmentation in medical images. DCM-Net significantly outperforms all baseline models in segmentation tasks, with higher Dice Similarity Coefficient (DSC) and mean Intersection over Union (mIoU) scores. Its robustness confirms strong potential for clinical use.

Medical imaging foundation models must adapt over time, yet full retraining is often blocked by privacy constraints and cost. We present a continual learning framework that avoids storing patient exemplars by pairing class conditional diffusion replay with Elastic Weight Consolidation. Using a compact Vision Transformer backbone, we evaluate across eight MedMNIST v2 tasks and CheXpert. On CheXpert our approach attains 0.851 AUROC, reduces forgetting by more than 30\% relative to DER\texttt{++}, and approaches joint training at 0.869 AUROC, while remaining efficient and privacy preserving. Analyses connect forgetting to two measurable factors: fidelity of replay and Fisher weighted parameter drift, highlighting the complementary roles of replay diffusion and synaptic stability. The results indicate a practical route for scalable, privacy aware continual adaptation of clinical imaging models.

Federated learning (FL) allows collaborative model training across healthcare sites without sharing sensitive patient data. However, real-world FL deployment is often hindered by complex operational challenges that demand substantial human efforts. This includes: (a) selecting appropriate clients (hospitals), (b) coordinating between the central server and clients, (c) client-level data pre-processing, (d) harmonizing non-standardized data and labels across clients, and (e) selecting FL algorithms based on user instructions and cross-client data characteristics. However, the existing FL works overlook these practical orchestration challenges. These operational bottlenecks motivate the need for autonomous, agent-driven FL systems, where intelligent agents at each hospital client and the central server agent collaboratively manage FL setup and model training with minimal human intervention. To this end, we first introduce an agent-driven FL framework that captures key phases of real-world FL workflows from client selection to training completion and a benchmark dubbed FedAgentBench that evaluates the ability of LLM agents to autonomously coordinate healthcare FL. Our framework incorporates 40 FL algorithms, each tailored to address diverse task-specific requirements and cross-client characteristics. Furthermore, we introduce a diverse set of complex tasks across 201 carefully curated datasets, simulating 6 modality-specific real-world healthcare environments, viz., Dermatoscopy, Ultrasound, Fundus, Histopathology, MRI, and X-Ray. We assess the agentic performance of 14 open-source and 10 proprietary LLMs spanning small, medium, and large model scales. While some agent cores such as GPT-4.1 and DeepSeek V3 can automate various stages of the FL pipeline, our results reveal that more complex, interdependent tasks based on implicit goals remain challenging for even the strongest models.

Federated Learning (FL) trains models locally at each research center or clinic and aggregates only model updates, making it a natural fit for medical imaging, where strict privacy laws forbid raw data sharing. A major obstacle is scanner-induced domain shift: non-biological variations in hardware or acquisition protocols can cause models to fail on external sites. Most harmonization methods correct this shift by directly comparing data across sites, conflicting with FL's privacy constraints. Domain Generalization (DG) offers a privacy-friendly alternative - learning site-invariant representations without sharing raw data - but standard DG pipelines still assume centralized access to multi-site data, again violating FL's guarantees. This paper meets these difficulties with a straightforward integration of a Domain-Adversarial Neural Network (DANN) within the FL process. After demonstrating that a naive federated DANN fails to converge, we propose a proximal regularization method that stabilizes adversarial training among clients. Experiments on T1-weighted 3-D brain MRIs from the OpenBHB dataset, performing brain-age prediction on participants aged 6-64 y (mean 22+/-6 y; 45 percent male) in training and 6-79 y (mean 19+/-13 y; 55 percent male) in validation, show that training on 15 sites and testing on 19 unseen sites yields superior cross-site generalization over FedAvg and ERM while preserving data privacy.

Automated vertebra analysis (AVA), encompassing vertebra detection and segmentation, plays a critical role in computer-aided diagnosis, surgical planning, and postoperative evaluation in spine-related clinical workflows. Despite notable progress, AVA continues to face key challenges, including variations in the field of view (FOV), complex vertebral morphology, limited availability of high-quality annotated data, and performance degradation under domain shifts. Over the past decade, numerous studies have employed deep learning (DL) to tackle these issues, introducing advanced network architectures and innovative learning paradigms. However, the rapid evolution of these methods has not been comprehensively captured by existing surveys, resulting in a knowledge gap regarding the current state of the field. To address this, this paper presents an up-to-date review that systematically summarizes recent advances. The review begins by consolidating publicly available datasets and evaluation metrics to support standardized benchmarking. Recent DL-based AVA approaches are then analyzed from two methodological perspectives: network architecture improvement and learning strategies design. Finally, an examination of persistent technical barriers and emerging clinical needs that are shaping future research directions is provided. These include multimodal learning, domain generalization, and the integration of foundation models. As the most current survey in the field, this review provides a comprehensive and structured synthesis aimed at guiding future research toward the development of robust, generalizable, and clinically deployable AVA systems in the era of intelligent medical imaging.

While deep learning has catalyzed breakthroughs across numerous domains, its broader adoption in clinical settings is inhibited by the costly and time-intensive nature of data acquisition and annotation. To further facilitate medical machine learning, we present an ultrasound dataset of 10,223 brightness-mode (B-mode) images consisting of sagittal slices of porcine spinal cords (N = 25) before and after a contusion injury. We additionally benchmark the performance metrics of several state-of-the-art object detection algorithms to localize the site of injury and semantic segmentation models to label the anatomy for comparison and creation of task-specific architectures. Finally, we evaluate the zero-shot generalization capabilities of the segmentation models on human ultrasound spinal cord images to determine whether training on our porcine dataset is sufficient for accurately interpreting human data. Our results show that the YOLOv8 detection model outperforms all evaluated models for injury localization, achieving a mean Average Precision (mAP50-95) score of 0.606. Segmentation metrics indicate that the DeepLabv3 segmentation model achieves the highest accuracy on unseen porcine anatomy, with a Mean Dice score of 0.587, while SAMed achieves the highest mean Dice score generalizing to human anatomy (0.445). To the best of our knowledge, this is the largest annotated dataset of spinal cord ultrasound images made publicly available to researchers and medical professionals, as well as the first public report of object detection and segmentation architectures to assess anatomical markers in the spinal cord for methodology development and clinical applications.

Multimodal datasets offer valuable support for developing Clinical Decision Support Systems (CDSS), which leverage predictive models to enhance clinicians' decision-making. In this observational study, we present a dataset of suspected Coronary Artery Disease (CAD) patients - called MultiD4CAD - comprising imaging and clinical data. The imaging data obtained from Coronary Computed Tomography Angiography (CCTA) includes epicardial (EAT) and pericoronary (PAT) adipose tissue segmentations. These metabolically active fat tissues play a key role in cardiovascular diseases. In addition, clinical data include a set of biomarkers recognized as CAD risk factors. The validated EAT and PAT segmentations make the dataset suitable for training predictive models based on radiomics and deep learning architectures. The inclusion of CAD disease labels allows for its application in supervised learning algorithms to predict CAD outcomes. MultiD4CAD has revealed important correlations between imaging features, clinical biomarkers, and CAD status. The article concludes by discussing some challenges, such as classification, segmentation, radiomics, and deep training tasks, that can be investigated and validated using the proposed dataset.

We present a patch-based 3D nnUNet adaptation for MR to CT and CBCT to CT image translation using the multicenter SynthRAD2025 dataset, covering head and neck (HN), thorax (TH), and abdomen (AB) regions. Our approach leverages two main network configurations: a standard UNet and a residual UNet, both adapted from nnUNet for image synthesis. The Anatomical Feature-Prioritized (AFP) loss was introduced, which compares multilayer features extracted from a compact segmentation network trained on TotalSegmentator labels, enhancing reconstruction of clinically relevant structures. Input volumes were normalized per-case using zscore normalization for MRIs, and clipping plus dataset level zscore normalization for CBCT and CT. Training used 3D patches tailored to each anatomical region without additional data augmentation. Models were trained for 1000 and 1500 epochs, with AFP fine-tuning performed for 500 epochs using a combined L1+AFP objective. During inference, overlapping patches were aggregated via mean averaging with step size of 0.3, and postprocessing included reverse zscore normalization. Both network configurations were applied across all regions, allowing consistent model design while capturing local adaptations through residual learning and AFP loss. Qualitative and quantitative evaluation revealed that residual networks combined with AFP yielded sharper reconstructions and improved anatomical fidelity, particularly for bone structures in MR to CT and lesions in CBCT to CT, while L1only networks achieved slightly better intensity-based metrics. This methodology provides a stable solution for cross modality medical image synthesis, demonstrating the effectiveness of combining the automatic nnUNet pipeline with residual learning and anatomically guided feature losses.

Recent advances in vision-language models (VLMs) have achieved remarkable performance on standard medical benchmarks, yet their true clinical reasoning ability remains unclear. Existing datasets predominantly emphasize classification accuracy, creating an evaluation illusion in which models appear proficient while still failing at high-stakes diagnostic reasoning. We introduce Neural-MedBench, a compact yet reasoning-intensive benchmark specifically designed to probe the limits of multimodal clinical reasoning in neurology. Neural-MedBench integrates multi-sequence MRI scans, structured electronic health records, and clinical notes, and encompasses three core task families: differential diagnosis, lesion recognition, and rationale generation. To ensure reliable evaluation, we develop a hybrid scoring pipeline that combines LLM-based graders, clinician validation, and semantic similarity metrics. Through systematic evaluation of state-of-the-art VLMs, including GPT-4o, Claude-4, and MedGemma, we observe a sharp performance drop compared to conventional datasets. Error analysis shows that reasoning failures, rather than perceptual errors, dominate model shortcomings. Our findings highlight the necessity of a Two-Axis Evaluation Framework: breadth-oriented large datasets for statistical generalization, and depth-oriented, compact benchmarks such as Neural-MedBench for reasoning fidelity. We release Neural-MedBench at https://neuromedbench.github.io/ as an open and extensible diagnostic testbed, which guides the expansion of future benchmarks and enables rigorous yet cost-effective assessment of clinically trustworthy AI.