Benchmarking ECG Foundational Models: A Reality Check Across Clinical Tasks
This work provides a reality check on the generalization of ECG foundation models for clinical applications, highlighting gaps in cardiac structure and outcome prediction.
The study benchmarked eight ECG foundation models across 26 clinical tasks using 12 public datasets, finding that three models outperformed supervised baselines in adult ECG interpretation, while ECG-CPC excelled in other categories and showed efficient scaling with dataset size.
The 12-lead electrocardiogram (ECG) is a long-standing diagnostic tool. Yet machine learning for ECG interpretation remains fragmented, often limited to narrow tasks or datasets. Foundation models promise broader adaptability, but their generalization across diverse ECG tasks is not well understood. We benchmarked eight ECG foundation models on 26 clinically relevant tasks using 12 public datasets comprising 1,650 regression and classification targets. Models were evaluated under fine-tuning and frozen settings, with scaling analyses across dataset sizes. Results show heterogeneous performance across domains: in the most widely studied domain, adult ECG interpretation, three foundation models consistently outperformed strong supervised baselines. In contrast, ECG-CPC, a compact structured state-space model pretrained on HEEDB, dominated other categories where most foundation models failed to surpass supervised learning. Foundation models also displayed distinct scaling behaviors with dataset size, which are critical for small-scale clinical applications. Overall, while foundation models show promise for adult ECG analysis, substantial gaps remain in cardiac structure, outcome prediction, and patient characterization. Notably, ECG-CPC's strong performance despite being orders of magnitude smaller and consuming minimal computational resources highlights untapped opportunities for advancing ECG foundation models.