Development of an automated, reliable, and clinically meaningful artificial intelligence (AI) tool for diagnosing cardiac disease from conventional cardiovascular magnetic resonance (CMR) images

2026-07-22

AI Summary

This study addresses the limitations of traditional cardiovascular magnetic resonance (CMR) image interpretation, which relies heavily on expert experience, and overcomes the high annotation costs and clinical deployment challenges faced by existing AI approaches. The authors propose an automated diagnostic framework that integrates locally deployed open-source large language models with multimodal CMR imaging—specifically cine and late gadolinium enhancement (LGE) sequences. The framework leverages large language models to extract labels from clinical reports and combines three vision foundation models (DINO, VST, and UMedPT) through a two-stage fine-tuning strategy and ensemble learning to enhance performance. Evaluated on an independent test set, the method achieves AUCs of 0.959 and 0.966 for hypertrophic cardiomyopathy and cardiac amyloidosis, respectively. The fully open-source implementation significantly improves diagnostic accuracy, robustness, and reproducibility.

Abstract

Aims: Cardiovascular magnetic resonance (CMR) imaging enables non-invasive assessment of myocardial structure, function, and pathology, but requires substantial experience in interpretation of CMR images that could be supported by artificial intelligence (AI)-based models. However, use of AI models for enhanced CMR reading is limited by labor-intensive data curation, suboptimal model performance, and unclear implementation pathways. Methods and results: We developed an automated data curation pipeline for CMR-based cardiovascular disease (CVD) diagnosis, integrating open-source locally-run large language models (LLMs) to extract diagnostic labels from narrative CMR reports and preprocessing multimodal imaging data, including cine and late-gadolinium-enhancement (LGE) CMR sequences. Three vision foundation models (DINO, VST, UMedPT) were fine-tuned across these modalities in a two-stage approach. The dataset comprised hypertrophic cardiomyopathy (HCM), dilated cardiomyopathy (DCM), ischemic cardiomyopathy (ICM), cardiac amyloidosis (CA), and normal controls (NOR). A total of 988 curated cases were randomly divided into 742 for training and 246 for validation. Fine-tuned AI-models achieved high discriminative diagnostic performance on an independent test set comprising 1067 patients , with individual AUC-ROC values of up to 0.937 for the correct diagnosis of HCM and 0.945 for cardiac amyloidosis. Ensemble strategies combining multiple models and modalities further improved AI-based diagnostic accuracy and robustness, achieving the highest overall diagnostic performance for HCM (AUC=0.959, CI [0.936-0.978]), CA (AUC=0.966, CI [0.939-0.986]), NOR (AUC=0.872, CI [0.852-0.894]), DCM (AUC=0.848, CI [0.808-0.885]) and ICM (AUC=0.840, CI [0.809-0.868]). All training and inference code, along with the trained model weights, are publicly available on https://github.com/sinaamirrajab/CMR_CVD.

Problem

• cardiovascular magnetic resonance
• artificial intelligence
• cardiac disease diagnosis
• automated data curation
• clinical implementation

Innovation

• automated data curation
• vision foundation models
• multimodal CMR
• ensemble AI diagnosis
• large language models