MMMuLNet: Multimodal Mutual Deep learning Framework for the Pediatric Congenital Heart Disease Detection.

The framework was validated on a dataset of 751 pediatric patients in Bangladesh, comprising 3,435 synchronized PCG-ECG recordings with expert-confirmed labels.

In 10-fold patient-wise cross validation, phonocardiograms alone achieved 90% accuracy, electrocardiograms alone 79%, and symptoms alone 73%.

Importantly, even under degraded signal conditions when both phonocardiograms and electrocardiograms were of low quality, the multimodal framework maintained an accuracy of 88% and an AUROC of 87%, demonstrating superior performance compared to the unimodal models.