EchoBridge: Long-Tail-Aware ECG-Echocardiography Text Alignment for Echocardiography-Derived Cardiac Findings
2026-07-27 • Machine Learning
Machine LearningArtificial Intelligence
AI summaryⓘ
The authors developed EchoBridge, a method to better learn heart-related information from ECG and echocardiography data together. Their approach separates shared and unique features of each type of data and adjusts how heart condition categories are represented to improve detection, especially for rare findings. They tested EchoBridge on multiple datasets and tasks, showing it outperforms other methods in accuracy without needing extra classifier training. Their method works well both within the same data source and when applied to new hospitals. Overall, it helps detect various heart problems, including uncommon valve conditions, more reliably.
echocardiographyECGrepresentation learningcross-modal alignmentprototype learningAUROCAUPRCvalvular heart diseaselinear probing
Authors
Xiaocheng Fang, Jieyi Cai, Guangkun Nie, Haoyu Wang, Jiarui Jin, Yujie Xiao, Bo Liu, Chenyang He, Qinghao Zhao, Gaofeng Cheng, Hongyan Li, Shenda Hong
Abstract
Standardized echocardiography conclusions provide meaningful supervision for learning ECG representations of echocardiography-derived cardiac findings. Global ECG--text alignment may entangle modality-specific factors, while long-tailed finding distributions provide sparse positive supervision for low-prevalence conditions. We propose EchoBridge with Complementary Shared--Private Projection (CSPP) and Adaptive Prototype Boundary Calibration (APBC). CSPP maps each modality into shared and auxiliary private projections, reduces directional redundancy via within-modality orthogonality, and bidirectionally aligns normalized shared projections. APBC organizes the shared hypersphere with class-specific prototypes, training-frequency-adaptive angular margins, and spherical Riesz repulsion. We evaluate EchoBridge on EchoNext-Mini and independent PKUPH and SHTMU cohorts under four protocols: prompt-based inference without downstream classifier training, in-domain frozen linear probing, target-domain cross-center frozen linear probing, and source-only cross-center transfer, supplemented by finding-specific analyses. EchoBridge improves classifier-free AUROC, AUPRC, and F1 over the strongest baselines by 7.88, 5.61, and 4.54 points, respectively, and achieves the highest point estimates across all in-domain and target-domain probing budgets and both source-only transfer cohorts. Finding-specific analyses show gains for most conditions, including several low-prevalence valvular findings.