Functional Magnetic Resonance Imaging (fMRI) is essential for studying brain function and diagnosing neurological disorders, but current analysis methods face reproducibility and transferability issues due to complex pre-processing and task-specific models. We introduce NeuroSTORM (Neuroimaging Foundation Model with Spatial-Temporal Optimized Representation Modeling), a generalizable framework that directly learns from 4D fMRI volumes and enables efficient knowledge transfer across diverse applications. NeuroSTORM is pre-trained on 28.65 million fMRI frames (>9,000 hours) from over 50,000 subjects across multiple centers and ages 5 to 100. Using a Mamba backbone and a shifted scanning strategy, it efficiently processes full 4D volumes. We also propose a spatial-temporal optimized pre-training approach and task-specific prompt tuning to improve transferability. NeuroSTORM outperforms existing methods across five tasks: age/gender prediction, phenotype prediction, disease diagnosis, fMRI-to-image retrieval, and task-based fMRI classification. It demonstrates strong clinical utility on datasets from hospitals in the U.S., South Korea, and Australia, achieving top performance in disease diagnosis and cognitive phenotype prediction. NeuroSTORM provides a standardized, open-source foundation model to improve reproducibility and transferability in fMRI-based clinical research.

Alzheimer's disease (AD) is a progressive and irreversible neurodegenerative disease. Early and precise diagnosis of AD is crucial for timely intervention and treatment planning to alleviate the progressive neurodegeneration. However, most existing methods rely on single-modality data, which contrasts with the multifaceted approach used by medical experts. While some deep learning approaches process multi-modal data, they are limited to specific tasks with a small set of input modalities and cannot handle arbitrary combinations. This highlights the need for a system that can address diverse AD-related tasks, process multi-modal or missing input, and integrate multiple advanced methods for improved performance. In this paper, we propose ADAgent, the first specialized AI agent for AD analysis, built on a large language model (LLM) to address user queries and support decision-making. ADAgent integrates a reasoning engine, specialized medical tools, and a collaborative outcome coordinator to facilitate multi-modal diagnosis and prognosis tasks in AD. Extensive experiments demonstrate that ADAgent outperforms SOTA methods, achieving significant improvements in accuracy, including a 2.7% increase in multi-modal diagnosis, a 0.7% improvement in multi-modal prognosis, and enhancements in MRI and PET diagnosis tasks.

IntroductionAccurately predicting Alzheimers Disease (AD) progression is useful for clinical care. The 2019 TADPOLE (The Alzheimers Disease Prediction Of Longitudinal Evolution) challenge evaluated 92 algorithms from 33 teams worldwide. Unlike typical clinical prediction studies, TADPOLE accommodates (1) variable number of observed timepoints across patients, (2) missing data across modalities and visits, and (3) prediction over an open-ended time horizon, which better reflects real-world data. However, TADPOLE only used the Alzheimers Disease Neuroimaging Initiative (ADNI) dataset, so how well top algorithms generalize to other cohorts remains unclear. MethodsWe tested five algorithms in three external datasets covering 2,312 participants and 13,200 timepoints. The algorithms included FROG, the overall TADPOLE winner, which utilized a unique Longitudinal-to-Cross-sectional (L2C) transformation to convert variable-length longitudinal histories into feature vectors of the same length across participants (i.e., same-length feature vectors). We also considered two FROG variants. One variant unified all XGBoost models from the original FROG with a single feedforward neural network (FNN), which we referred to as L2C-FNN. We also included minimal recurrent neural networks (MinimalRNN), which was ranked second at publication time, as well as AD Course Map (AD-Map), which outperformed MinimalRNN at publication time. All five models - three FROG variants, MinimalRNN and AD-Map - were trained on ADNI and tested on the external datasets. ResultsL2C-FNN performed the best overall. In the case of predicting cognition and ventricle volume, L2C-FNN and AD-Map were the best. For clinical diagnosis prediction, L2C-FNN was the best, while AD-Map was the worst. L2C-FNN also maintained its edge over other models, regardless of the number of observed timepoints, and regardless of the prediction horizon from 0 to 6 years into the future. ConclusionsL2C-FNN shows strong potential for both short-term and long-term dementia progression prediction. Pretrained ADNI models are available: https://github.com/ThomasYeoLab/CBIG/tree/master/stable_projects/predict_phenotypes/Zhang2025_L2CFNN.

Small cell lung cancer (SCLC) is aggressive with poor survival outcomes, and most patients develop resistance to chemotherapy. No predictive biomarkers currently guide therapy. This study evaluates radiomic features to predict PFS and OS in limited-stage SCLC (LS-SCLC) and assesses PFS, OS, and the added benefit of chemoimmunotherapy (CHIO) in extensive-stage SCLC (ES-SCLC). A total of 660 SCLC patients (470 ES-SCLC, 190 LS-SCLC) from three sites were analyzed. LS-SCLC patients received chemotherapy and radiation, while ES-SCLC patients received either chemotherapy alone or chemoimmunotherapy. Radiomic and quantitative vasculature tortuosity features were extracted from CT scans. A LASSO-Cox regression model was used to construct the ES- Risk-Score (ESRS) and LS- Risk-Score (LSRS). ESRS was associated with PFS in training (HR = 1.54, adj. P = .0013) and validation sets (HR = 1.32, adj. P = .0001; HR = 2.4, adj. P = .0073) and with OS in training (HR = 1.37, adj. P = .0054) and validation sets (HR = 1.35, adj. P < .0006; HR = 1.6, adj. P < .0085) in ES-SCLC patients treated with chemotherapy. High-risk patients had improved PFS (HR = 0.68, adj. P < .001) and OS (HR = 0.78, adj. P = .026) with chemoimmunotherapy. LSRS was associated with PFS in training and validation sets (HR = 1.9, adj. P = .007; HR = 1.4, adj. P = .0098; HR = 2.1, adj. P = .028) in LS-SCLC patients receiving chemoradiation. Radiomics is prognostic for PFS and OS and predicts chemoimmunotherapy benefit in high-risk ES-SCLC patients.

Speed-of-sound (SoS) is a biomechanical characteristic of tissue, and its imaging can provide a promising biomarker for diagnosis. Reconstructing SoS images from ultrasound acquisitions can be cast as a limited-angle computed-tomography problem, with variational networks being a promising model-based deep learning solution. Some acquired data frames may, however, get corrupted by noise due to, e.g., motion, lack of contact, and acoustic shadows, which in turn negatively affects the resulting SoS reconstructions. We propose to use the uncertainty in SoS reconstructions to attribute trust to each individual acquired frame. Given multiple acquisitions, we then use an uncertainty-based automatic selection among these retrospectively, to improve diagnostic decisions. We investigate uncertainty estimation based on Monte Carlo Dropout and Bayesian Variational Inference. We assess our automatic frame selection method for differential diagnosis of breast cancer, distinguishing between benign fibroadenoma and malignant carcinoma. We evaluate 21 lesions classified as BI-RADS 4, which represents suspicious cases for probable malignancy. The most trustworthy frame among four acquisitions of each lesion was identified using uncertainty-based criteria. Selecting a frame informed by uncertainty achieved an area under curve of 76% and 80% for Monte Carlo Dropout and Bayesian Variational Inference, respectively, superior to any uncertainty-uninformed baselines with the best one achieving 64%. A novel use of uncertainty estimation is proposed for selecting one of multiple data acquisitions for further processing and decision making.

Chronic obstructive pulmonary disease (COPD) is one of the leading causes of morbidity and mortality in China, highlighting the importance of early diagnosis and ongoing monitoring for effective management. In recent years, hyperpolarized 129Xe MRI technology has gained significant clinical attention due to its ability to non-invasively and visually assess lung ventilation, microstructure, and gas exchange function. Its recent clinical approval in China, the United States and several European countries, represents a significant advancement in pulmonary imaging. This review provides an overview of the latest developments in hyperpolarized 129Xe MRI technology for COPD assessment in China. It covers the progress in instrument development, advanced imaging techniques, artificial intelligence-driven reconstruction methods, molecular imaging, and the application of this technology in both COPD patients and animal models. Furthermore, the review explores potential technical innovations in 129Xe MRI and discusses future directions for its clinical applications, aiming to address existing challenges and expand the technology's impact in clinical practice.

Abdominal aortic aneurysms (AAAs) are progressive focal dilatations of the abdominal aorta. AAAs may rupture, with a survival rate of only 20\%. Current clinical guidelines recommend elective surgical repair when the maximum AAA diameter exceeds 55 mm in men or 50 mm in women. Patients that do not meet these criteria are periodically monitored, with surveillance intervals based on the maximum AAA diameter. However, this diameter does not take into account the complex relation between the 3D AAA shape and its growth, making standardized intervals potentially unfit. Personalized AAA growth predictions could improve monitoring strategies. We propose to use an SE(3)-symmetric transformer model to predict AAA growth directly on the vascular model surface enriched with local, multi-physical features. In contrast to other works which have parameterized the AAA shape, this representation preserves the vascular surface's anatomical structure and geometric fidelity. We train our model using a longitudinal dataset of 113 computed tomography angiography (CTA) scans of 24 AAA patients at irregularly sampled intervals. After training, our model predicts AAA growth to the next scan moment with a median diameter error of 1.18 mm. We further demonstrate our model's utility to identify whether a patient will become eligible for elective repair within two years (acc = 0.93). Finally, we evaluate our model's generalization on an external validation set consisting of 25 CTAs from 7 AAA patients from a different hospital. Our results show that local directional AAA growth prediction from the vascular surface is feasible and may contribute to personalized surveillance strategies.

Mesenteric malperfusion (MMP) is an uncommon but devastating complication of acute aortic dissection (AAD) that combines 2 life-threatening conditions-aortic dissection and acute mesenteric ischemia. The complex pathophysiology of MMP poses substantial diagnostic and management challenges. Currently, delayed diagnosis remains a critical contributor to poor outcomes because of the absence of reliable individualized risk assessment tools. This study aims to develop and validate a deep learning-based model that integrates multimodal data to identify patients with AAD at high risk of MMP. This multicenter retrospective study included 525 patients with AAD from 2 hospitals. The training and internal validation cohort consisted of 450 patients from Beijing Anzhen Hospital, whereas the external validation cohort comprised 75 patients from Nanjing Drum Tower Hospital. Three machine learning models were developed: the benchmark model using laboratory parameters, the multiorgan feature-based AAD complicating MMP (MAM) model based on computed tomography angiography images, and the integrated model combining both data modalities. Model performance was assessed using the area under the curve, accuracy, sensitivity, specificity, and Brier score. To improve interpretability, gradient-weighted class activation mapping was used to identify and visualize discriminative imaging features. Univariate and multivariate regression analyses were used to evaluate the prognostic significance of the risk score generated by the optimal model. In the external validation cohort, the integrated model demonstrated superior performance, with an area under the curve of 0.780 (95% CI 0.777-0.785), which was significantly greater than those of the benchmark model (0.586, 95% CI 0.574-0.586) and the MAM model (0.732, 95% CI 0.724-0.734). This highlights the benefits of multimodal integration over single-modality approaches. Additional classification metrics revealed that the integrated model had an accuracy of 0.760 (95% CI 0.758-0.764), a sensitivity of 0.667 (95% CI 0.659-0.675), a specificity of 0.783 (95% CI 0.781-0.788), and a Brier score of 0.143 (95% CI 0.143-0.145). Moreover, gradient-weighted class activation mapping visualizations of the MAM model revealed that during positive predictions, the model focused more on key anatomical areas, particularly the superior mesenteric artery origin and intestinal regions with characteristic gas or fluid accumulation. Univariate and multivariate analyses also revealed that the risk score derived from the integrated model was independently associated with inhospital mortality risk among patients with AAD undergoing endovascular or surgical treatment (odds ratio 1.030, 95% CI 1.004-1.056; P=.02). Our findings demonstrate that compared with unimodal approaches, an integrated deep learning model incorporating both imaging and clinical data has greater diagnostic accuracy for MMP in patients with AAD. This model may serve as a valuable tool for early risk identification, facilitating timely therapeutic decision-making. Further prospective validation is warranted to confirm its clinical utility. Chinese Clinical Registry Center ChiCTR2400086050; http://www.chictr.org.cn/showproj.html?proj=226129.

Foundation models (FMs) are changing the way medical images are analyzed by learning from large collections of unlabeled data. Instead of relying on manually annotated examples, FMs are pre-trained to learn general-purpose visual features that can later be adapted to specific clinical tasks with little additional supervision. In this review, we examine how FMs are being developed and applied in pathology, radiology, and ophthalmology, drawing on evidence from over 150 studies. We explain the core components of FM pipelines, including model architectures, self-supervised learning methods, and strategies for downstream adaptation. We also review how FMs are being used in each imaging domain and compare design choices across applications. Finally, we discuss key challenges and open questions to guide future research.

It is not rare to experience a biochemical recurrence (BCR) following radical prostatectomy (RP) for prostate cancer (PCa). It has been reported that early detection and management of BCR following surgery could improve survival in PCa. This study aimed to develop a nomogram integrating deep learning-based radiomic features and clinical parameters to predict 1-year BCR after RP and to examine the associations between radiomic scores and the tumor microenvironment (TME). In this retrospective multicenter study, two independent cohorts of patients (n = 349) who underwent RP after multiparametric magnetic resonance imaging (mpMRI) between January 2015 and January 2022 were included in the analysis. Single-cell RNA sequencing data from four prospectively enrolled participants were used to investigate the radiomic score-related TME. The 3D U-Net was trained and optimized for prostate cancer segmentation using diffusion-weighted imaging, and radiomic features of the target lesion were extracted. Predictive nomograms were developed via multivariate Cox proportional hazard regression analysis. The nomograms were assessed for discrimination, calibration, and clinical usefulness. In the development cohort, the clinical-radiomic nomogram had an AUC of 0.892 (95% confidence interval: 0.783--0.939), which was considerably greater than those of the radiomic signature and clinical model. The Hosmer-Lemeshow test demonstrated that the clinical-radiomic model performed well in both the development (P = 0.461) and validation (P = 0.722) cohorts. Decision curve analysis revealed that the clinical-radiomic nomogram displayed better clinical predictive usefulness than the clinical or radiomic signature alone in both cohorts. Radiomic scores were associated with a significant difference in the TME pattern. Our study demonstrated the feasibility of a DWI-based clinical-radiomic nomogram combined with deep learning for the prediction of 1-year BCR. The findings revealed that the radiomic score was associated with a distinctive tumor microenvironment.