PaperarXiv:2611.12307
PACE-ECG: Patient-Aware Contrastive ECG Pretraining; Leveraging Longitudinal Structure for Transferable Representations
arXiv preprint (cs.AI)
(opens in a new tab)Cavarra, Riccardo et al., "PACE-ECG: Patient-Aware Contrastive ECG Pretraining; Leveraging Longitudinal Structure for Transferable Representations", arXiv:2611.12307, CC BY 4.0
TL;DR
King's College London ECG pretraining that pairs temporally neighbouring ECGs from the same patient (plus clinically guided multitask heads). Frozen 41x-smaller encoder beats DeepECG-SSL on post-MI mortality and incident heart failure, by +5.6 to +11.0 AUROC points on external HEEDB cohorts. No code or weights stated.
Abstract
Foundation models for physiological signals are commonly evaluated on in-domain interpretation tasks, but their ability to transfer to longitudinal clinical prognosis across healthcare systems remains uncertain. We developed PACE-ECG, an ECG pretraining framework that combines contrastive learning across temporally neighbouring recordings from the same patient with clinically guided multitask supervision. We evaluated frozen ECG representations for post-myocardial-infarction all-cause mortality and incident heart failure prediction across MIMIC-IV and two external HEEDB cohorts, comparing the proposed representation with DeepECG-SSL, a state-of-the-art model, and two control baselines. Across both downstream tasks, the proposed model outperformed DeepECG-SSL in-domain (AUROC +1.9% for mortality and +4.0% for heart failure) and on external populations in HEEDB-Boston (+5.6% for mortality +10.3% for heart failure) and HEEDB-Atlanta (+6.9% for mortality +11.0% for heart failure) while using an encoder 41 times smaller and a fraction of the pretraining data. A matched ablation showed that patient-temporal pairing improved external mortality and heart failure AUROC over generic self-supervision by 7.0% and 9.4% on HEEDB-Boston and 8.9% and 9.3% on HEEDB-Atlanta. These findings indicate that pretraining-objective design greatly influences downstream transfer alongside model and data scale, and that in-domain performance alone is insufficient to establish the clinical generalisability of ECG foundation representations.
Verbatim from arXiv (opens in a new tab) · arXiv metadata (abstracts are released under CC0)
Results on ChatPPG leaderboards
All leaderboardsNo results from this paper on ChatPPG leaderboards yet. Browse the leaderboards.
Tasks & signals
Signals
Keywords
Datasets used
How it works
Cite this paper
Cavarra, R., Lovatelli, L., Ogbomo-Harmitt, S., Grzelak, J., De Vecchi, A., King, A., & Aslanidi, O. (2026). PACE-ECG: Patient-Aware Contrastive ECG Pretraining; Leveraging Longitudinal Structure for Transferable Representations. arXiv preprint. https://arxiv.org/abs/2611.12307
Cavarra, Riccardo, Lupo Lovatelli, Shaheim Ogbomo-Harmitt, Jakub Grzelak, Adelaide De Vecchi, Andrew King, and Oleg Aslanidi. 2026. "PACE-ECG: Patient-Aware Contrastive ECG Pretraining; Leveraging Longitudinal Structure for Transferable Representations." arXiv preprint. https://arxiv.org/abs/2611.12307.
Cavarra, R., Lovatelli, L., Ogbomo-Harmitt, S., Grzelak, J., De Vecchi, A., King, A. and Aslanidi, O. (2026) 'PACE-ECG: Patient-Aware Contrastive ECG Pretraining; Leveraging Longitudinal Structure for Transferable Representations', arXiv preprint. Available at: https://arxiv.org/abs/2611.12307
Cavarra, Riccardo, et al. "PACE-ECG: Patient-Aware Contrastive ECG Pretraining; Leveraging Longitudinal Structure for Transferable Representations." arXiv preprint, 2026. https://arxiv.org/abs/2611.12307.
R. Cavarra et al., "PACE-ECG: Patient-Aware Contrastive ECG Pretraining; Leveraging Longitudinal Structure for Transferable Representations," arXiv preprint, 2026, [Online]. Available: https://arxiv.org/abs/2611.12307.
@misc{cavarra2026paceecg,
title = {{PACE-ECG: Patient-Aware Contrastive ECG Pretraining; Leveraging Longitudinal Structure for Transferable Representations}},
author = {Cavarra, Riccardo and Lovatelli, Lupo and Ogbomo-Harmitt, Shaheim and Grzelak, Jakub and De Vecchi, Adelaide and King, Andrew and Aslanidi, Oleg},
year = {2026},
howpublished = {arXiv preprint},
eprint = {2611.12307},
archivePrefix = {arXiv}
}Added to the ChatPPG research hub
