AI-Powered Early Detection and Prognostic Modeling of Restrictive Cardiomyopathy Using Multimodal Non-Invasive Data

Restrictive Cardiomyopathy (RCM) is a rare but severe heart disease that is often diagnosed in advanced stages, leading to significant clinical consequences. Detecting RCM at an early stage is essential to slowing disease progression and improving patient outcomes. This study introduces a novel approach that leverages multimodal non-invasive data, including electronic health records (EHRs), medical imaging, and genetic information, to enhance early detection and prognosis. The model underwent rigorous training and validation using the ACDC, MIMIC-IV, ClinVar datasets, employing deep learning techniques for feature extraction and classification. The system demonstrated high accuracy (93%), precision (0.90), and recall (0.91%), surpassing conventional diagnostic methods. By analyzing longitudinal patient data, the proposed method identifies subtle biomarkers and predictive patterns indicative of RCM onset. Additionally, it provides personalized prognostic insights, such as assessing the likelihood of heart failure or arrhythmias, all while seamlessly integrating into existing clinical workflows without requiring additional hardware. This research contributes to the advancement of cardiology by incorporating AI-driven methodologies that improve diagnostic accuracy and enhance patient-centered care.