Foundation models have demonstrated remarkable potential in medical domain. However, their application to complex cardiovascular diagnostics remains underexplored. In this paper, we present Cardiac-CLIP, a multi-modal foundation model designed for 3D cardiac CT images. Cardiac-CLIP is developed through a two-stage pre-training strategy. The first stage employs a 3D masked autoencoder (MAE) to perform self-supervised representation learning from large-scale unlabeled volumetric data, enabling the visual encoder to capture rich anatomical and contextual features. In the second stage, contrastive learning is introduced to align visual and textual representations, facilitating cross-modal understanding. To support the pre-training, we collect 16641 real clinical CT scans, supplemented by 114k publicly available data. Meanwhile, we standardize free-text radiology reports into unified templates and construct the pathology vectors according to diagnostic attributes, based on which the soft-label matrix is generated to supervise the contrastive learning process. On the other hand, to comprehensively evaluate the effectiveness of Cardiac-CLIP, we collect 6,722 real-clinical data from 12 independent institutions, along with the open-source data to construct the evaluation dataset. Specifically, Cardiac-CLIP is comprehensively evaluated across multiple tasks, including cardiovascular abnormality classification, information retrieval and clinical analysis. Experimental results demonstrate that Cardiac-CLIP achieves state-of-the-art performance across various downstream tasks in both internal and external data. Particularly, Cardiac-CLIP exhibits great effectiveness in supporting complex clinical tasks such as the prospective prediction of acute coronary syndrome, which is notoriously difficult in real-world scenarios.

Ultrasound is widely used in clinical care, yet standard deep learning methods often struggle with full video analysis due to non-standardized acquisition and operator bias. We offer a new perspective on ultrasound video analysis through implicit neural representations (INRs). We build on Functa, an INR framework in which each image is represented by a modulation vector that conditions a shared neural network. However, its extension to the temporal domain of medical videos remains unexplored. To address this gap, we propose VidFuncta, a novel framework that leverages Functa to encode variable-length ultrasound videos into compact, time-resolved representations. VidFuncta disentangles each video into a static video-specific vector and a sequence of time-dependent modulation vectors, capturing both temporal dynamics and dataset-level redundancies. Our method outperforms 2D and 3D baselines on video reconstruction and enables downstream tasks to directly operate on the learned 1D modulation vectors. We validate VidFuncta on three public ultrasound video datasets -- cardiac, lung, and breast -- and evaluate its downstream performance on ejection fraction prediction, B-line detection, and breast lesion classification. These results highlight the potential of VidFuncta as a generalizable and efficient representation framework for ultrasound videos. Our code is publicly available under https://github.com/JuliaWolleb/VidFuncta_public.

Recent deep learning approaches have shown promise in learning such individual brain parcellations from functional magnetic resonance imaging (fMRI). However, most existing methods assume consistent data distributions across domains and struggle with domain shifts inherent to real-world cross-dataset scenarios. To address this challenge, we proposed Graph Domain Adaptation for Individual Parcellation (GDAIP), a novel framework that integrates Graph Attention Networks (GAT) with Minimax Entropy (MME)-based domain adaptation. We construct cross-dataset brain graphs at both the group and individual levels. By leveraging semi-supervised training and adversarial optimization of the prediction entropy on unlabeled vertices from target brain graph, the reference atlas is adapted from the group-level brain graph to the individual brain graph, enabling individual parcellation under cross-dataset settings. We evaluated our method using parcellation visualization, Dice coefficient, and functional homogeneity. Experimental results demonstrate that GDAIP produces individual parcellations with topologically plausible boundaries, strong cross-session consistency, and ability of reflecting functional organization.

Medical image-language pre-training aims to align medical images with clinically relevant text to improve model performance on various downstream tasks. However, existing models often struggle with the variability and ambiguity inherent in medical data, limiting their ability to capture nuanced clinical information and uncertainty. This work introduces an uncertainty-aware medical image-text pre-training model that enhances generalization capabilities in medical image analysis. Building on previous methods and focusing on Chest X-Rays, our approach utilizes structured text reports generated by a large language model (LLM) to augment image data with clinically relevant context. These reports begin with a definition of the disease, followed by the `appearance' section to highlight critical regions of interest, and finally `observations' and `verdicts' that ground model predictions in clinical semantics. By modeling both inter- and intra-modal uncertainty, our framework captures the inherent ambiguity in medical images and text, yielding improved representations and performance on downstream tasks. Our model demonstrates significant advances in medical image-text pre-training, obtaining state-of-the-art performance on multiple downstream tasks.

Accurate segmentation of the prostate peripheral zone (PZ) in T2-weighted MRI is critical for the early detection of prostate cancer. Existing segmentation methods are hindered by significant inter-observer variability (37.4 ± 5.6%), poor boundary localization, and the presence of motion artifacts, along with challenges in clinical integration. In this study, we propose BioAug-Net, a novel framework that integrates real-time physiological signal feedback with MRI data, leveraging transformer-based attention mechanisms and a probabilistic clinical decision support system (PCDSS). BioAug-Net features a dual-branch asymmetric attention mechanism: one branch processes spatial MRI features, while the other incorporates temporal sensor signals through a BiGRU-driven adaptive masking module. Additionally, a Markov Decision Process-based PCDSS maps segmentation outputs to clinical PI-RADS scores, with uncertainty quantification. We validated BioAug-Net on a multi-institutional dataset (n=1,542) and demonstrated state-of-the-art performance, achieving a Dice Similarity Coefficient of 89.7% (p < 0.001), sensitivity of 91.2% (p < 0.001), specificity of 88.4% (p < 0.001), and HD95 of 2.14 mm (p < 0.001), outperforming U-Net, Attention U-Net, and TransUNet. Sensor integration improved segmentation accuracy by 12.6% (p < 0.001) and reduced inter-observer variation by 48.3% (p < 0.001). Radiologist evaluations (n=3) confirmed a 15.0% reduction in diagnosis time (p = 0.003) and an increase in inter-reader agreement from K = 0.68 to K = 0.82 (p = 0.001). Our results show that BioAug-Net offers a clinically viable solution for early prostate cancer detection through enhanced physiological coupling and explainable AI diagnostics.

High-grade gliomas are highly invasive and respond variably to chemoradiation. Accurate, patient-specific predictions of tumor response could enhance treatment planning. We present a novel computational platform that assimilates MRI data to continually predict spatiotemporal tumor changes during chemoradiotherapy. Tumor growth and response to chemoradiation was described using a two-species reaction-diffusion model of enhancing and non-enhancing regions of the tumor. Two evaluation scenarios were used to test the predictive accuracy of this model. In scenario 1, the model was calibrated on a patient-specific basis (n = 21) to weekly MRI data during the course of chemoradiotherapy. A data assimilation framework was used to update model parameters with each new imaging visit which were then used to update model predictions. In scenario 2, we evaluated the predictive accuracy of the model when fewer data points are available by calibrating the same model using only the first two imaging visits and then predicted tumor response at the remaining five weeks of treatment. We investigated three approaches to assign model parameters for scenario 2: (1) predictions using only parameters estimated by fitting the data obtained from an individual patient's first two imaging visits, (2) predictions made by averaging the patient-specific parameters with the cohort-derived parameters, and (3) predictions using only cohort-derived parameters. Scenario 1 achieved a median [range] concordance correlation coefficient (CCC) between the predicted and measured total tumor cell counts of 0.91 [0.84, 0.95], and a median [range] percent error in tumor volume of -2.6% [-19.7, 8.0%], demonstrating strong agreement throughout the course of treatment. For scenario 2, the three approaches yielded CCCs of: (1) 0.65 [0.51, 0.88], (2) 0.74 [0.70, 0.91], (3) 0.76 [0.73, 0.92] with significant differences between the approach (1) that does not use the cohort parameters and the two approaches (2 and 3) that do. The proposed data assimilation framework enhances the accuracy of tumor growth forecasts by integrating patient-specific and cohort-based data. These findings show a practical method for identifying more personalized treatment strategies in high-grade glioma patients.

The Artificial Intelligence in Medical Imaging (AIMI) initiative aims to enhance the National Cancer Institute's (NCI) Image Data Commons (IDC) by releasing fully reproducible nnU-Net models, along with AI-assisted segmentation for cancer radiology images. In this extension of our earlier work, we created high-quality, AI-annotated imaging datasets for 11 IDC collections, spanning computed tomography (CT) and magnetic resonance imaging (MRI) of the lungs, breast, brain, kidneys, prostate, and liver. Each nnU-Net model was trained on open-source datasets, and a portion of the AI-generated annotations was reviewed and corrected by board-certified radiologists. Both the AI and radiologist annotations were encoded in compliance with the Digital Imaging and Communications in Medicine (DICOM) standard, ensuring seamless integration into the IDC collections. By making these models, images, and annotations publicly accessible, we aim to facilitate further research and development in cancer imaging.

Deep learning (DL) demonstrates high sensitivity but low specificity in lung cancer (LC) detection during CT screening, and the seven Tumor-associated antigens autoantibodies (7-TAAbs), known for its high specificity in LC, was employed to improve the DL's specificity for the efficiency of LC screening in China. To develop and evaluate a risk model combining 7-TAAbs test and DL scores for diagnosing LC with pulmonary lesions < 70 mm. Four hundreds and six patients with 406 lesions were enrolled and assigned into training set (n = 313) and test set (n = 93) randomly. The malignant lesions were defined as those lesions with high malignant risks by DL or those with positive expression of 7-TAAbs panel. Model performance was assessed using the area under the receiver operating characteristic curves (AUC). In the training set, the AUCs for DL, 7-TAAbs, combined model (DL and 7-TAAbs) and combined model (DL or 7-TAAbs) were 0.771, 0.638, 0.606, 0.809 seperately. In the test set, the combined model (DL or 7-TAAbs) achieved achieved the highest sensitivity (82.6%), NPV (81.8%) and accuracy (79.6%) among four models, and the AUCs of DL model, 7-TAAbs model, combined model (DL and 7-TAAbs), and combined model (DL or 7-TAAbs) were 0.731, 0.679, 0.574, and 0.794, respectively. The 7-TAAbs test significantly enhances DL performance in predicting LC with pulmonary leisons < 70 mm in China.

Physiological brain aging is associated with cognitive impairment and neuroanatomical changes. Brain age prediction of routine clinical 2D brain MRI scans were understudied and often unsuccessful. We developed a novel brain age prediction framework for clinical 2D T1-weighted MRI scans using a deep learning-based model trained with research grade 3D MRI scans mostly from publicly available datasets (N = 8681; age = 51.76 ± 21.74). Our model showed accurate and fast brain age prediction on clinical 2D MRI scans from cognitively unimpaired (CU) subjects (N = 175) with MAE of 2.73 years after age bias correction (Pearson's r = 0.918). Brain age gap of Alzheimer's disease (AD) subjects was significantly greater than CU subjects (p < 0.001) and increase in brain age gap was associated with disease progression in both AD (p < 0.05) and Parkinson's disease (p < 0.01). Our framework can be extended to other MRI modalities and potentially applied to routine clinical examinations, enabling early detection of structural anomalies and improve patient outcome.

Alzheimer's disease is a progressive neurological disorder that profoundly affects cognitive functions and daily activities. Rapid and precise identification is essential for effective intervention and improved patient outcomes. This research introduces an innovative hybrid filtering approach with a deep transfer learning model for detecting Alzheimer's disease utilizing brain imaging data. The hybrid filtering method integrates the Adaptive Non-Local Means filter with a Sharpening filter for image preprocessing. Furthermore, the deep learning model used in this study is constructed on the EfficientNetV2B3 architecture, augmented with additional layers and fine-tuning to guarantee effective classification among four categories: Mild, moderate, very mild, and non-demented. The work employs Grad-CAM++ to enhance interpretability by localizing disease-relevant characteristics in brain images. The experimental assessment, performed on a publicly accessible dataset, illustrates the ability of the model to achieve an accuracy of 99.45%. These findings underscore the capability of sophisticated deep learning methodologies to aid clinicians in accurately identifying Alzheimer's disease.