ECGFlowCMR: Pretraining with ECG-Generated Cine CMR Improves Cardiac Disease Classification and Phenotype Prediction
This addresses the challenge of high-cost cardiac imaging for medical diagnosis by enabling scalable pretraining, though it is incremental as it builds on existing generative and cross-modal methods.
The paper tackled the problem of limited labeled cardiac MRI data by proposing ECGFlowCMR, a framework that generates cine cardiac MRI sequences from electrocardiograms, which improved cardiac disease classification and phenotype prediction, achieving performance gains on datasets like the UK Biobank.
Cardiac Magnetic Resonance (CMR) imaging provides a comprehensive assessment of cardiac structure and function but remains constrained by high acquisition costs and reliance on expert annotations, limiting the availability of large-scale labeled datasets. In contrast, electrocardiograms (ECGs) are inexpensive, widely accessible, and offer a promising modality for conditioning the generative synthesis of cine CMR. To this end, we propose ECGFlowCMR, a novel ECG-to-CMR generative framework that integrates a Phase-Aware Masked Autoencoder (PA-MAE) and an Anatomy-Motion Disentangled Flow (AMDF) to address two fundamental challenges: (1) the cross-modal temporal mismatch between multi-beat ECG recordings and single-cycle CMR sequences, and (2) the anatomical observability gap due to the limited structural information inherent in ECGs. Extensive experiments on the UK Biobank and a proprietary clinical dataset demonstrate that ECGFlowCMR can generate realistic cine CMR sequences from ECG inputs, enabling scalable pretraining and improving performance on downstream cardiac disease classification and phenotype prediction tasks.