ECGFlowCMR: Pretraining with ECG-Generated Cine CMR Helps Cardiac Disease Classification and Phenotype Prediction

TL;DR

ECGFlowCMR introduces a generative framework that uses ECG data to synthesize cine CMR images, enhancing cardiac disease classification and phenotype prediction while reducing reliance on costly labeled CMR datasets.

Contribution

The paper presents a novel ECG-to-CMR generative model combining PA-MAE and AMDF to address cross-modal and structural challenges, enabling scalable pretraining for cardiac analysis.

Findings

Generated realistic cine CMR sequences from ECG inputs.

Improved performance on disease classification tasks.

Demonstrated effectiveness on UK Biobank and clinical datasets.

Abstract

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.