Beyond Machine Interpretation: Learning from Expert Over-Reads Improves ECG Diagnosis

Sunwoo Kwak, Fengbei Liu, Nusrat B. Nizam, Ilan Richter, Nir Uriel, Peter M. Okin, Mert R. Sabuncu
Proceedings of The 9th International Conference on Medical Imaging with Deep Learning, PMLR 315:36-55, 2026.

Abstract

Automated machine-read ECG interpretations are widely used in clinical practice but often unreliable, leading to systematic diagnostic errors. This work investigates how training with cardiologist over-reads impacts model accuracy and clinical reliability. Using a large paired corpus of over two million ECGs containing both machine and expert interpretations, we evaluate three learning paradigms: (i) supervised learning on expert over-read labels, (ii) Self-training that extends expert supervision to public ECGs, and (iii) multimodal contrastive learning with CLIP and NegCLIP. Across all settings, models trained with expert over-read data consistently outperform those trained on machine-read labels, especially for rare but clinically important conditions. Self-training and NegCLIP further demonstrate scalable strategies to propagate expert knowledge beyond labeled datasets. These findings highlight the essential role of expert over-reads in developing trustworthy and clinically aligned ECG AI systems.

Cite this Paper

BibTeX
@InProceedings{pmlr-v315-kwak26a, title = {Beyond Machine Interpretation: Learning from Expert Over-Reads Improves ECG Diagnosis}, author = {Kwak, Sunwoo and Liu, Fengbei and Nizam, Nusrat B. and Richter, Ilan and Uriel, Nir and Okin, Peter M. and Sabuncu, Mert R.}, booktitle = {Proceedings of The 9th International Conference on Medical Imaging with Deep Learning}, pages = {36--55}, year = {2026}, editor = {Huo, Yuankai and Gao, Mingchen and Kuo, Chang-Fu and Jin, Yueming and Deng, Ruining}, volume = {315}, series = {Proceedings of Machine Learning Research}, month = {08--10 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v315/main/assets/kwak26a/kwak26a.pdf}, url = {https://proceedings.mlr.press/v315/kwak26a.html}, abstract = {Automated machine-read ECG interpretations are widely used in clinical practice but often unreliable, leading to systematic diagnostic errors. This work investigates how training with cardiologist over-reads impacts model accuracy and clinical reliability. Using a large paired corpus of over two million ECGs containing both machine and expert interpretations, we evaluate three learning paradigms: (i) supervised learning on expert over-read labels, (ii) Self-training that extends expert supervision to public ECGs, and (iii) multimodal contrastive learning with CLIP and NegCLIP. Across all settings, models trained with expert over-read data consistently outperform those trained on machine-read labels, especially for rare but clinically important conditions. Self-training and NegCLIP further demonstrate scalable strategies to propagate expert knowledge beyond labeled datasets. These findings highlight the essential role of expert over-reads in developing trustworthy and clinically aligned ECG AI systems.} }
Endnote
%0 Conference Paper %T Beyond Machine Interpretation: Learning from Expert Over-Reads Improves ECG Diagnosis %A Sunwoo Kwak %A Fengbei Liu %A Nusrat B. Nizam %A Ilan Richter %A Nir Uriel %A Peter M. Okin %A Mert R. Sabuncu %B Proceedings of The 9th International Conference on Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2026 %E Yuankai Huo %E Mingchen Gao %E Chang-Fu Kuo %E Yueming Jin %E Ruining Deng %F pmlr-v315-kwak26a %I PMLR %P 36--55 %U https://proceedings.mlr.press/v315/kwak26a.html %V 315 %X Automated machine-read ECG interpretations are widely used in clinical practice but often unreliable, leading to systematic diagnostic errors. This work investigates how training with cardiologist over-reads impacts model accuracy and clinical reliability. Using a large paired corpus of over two million ECGs containing both machine and expert interpretations, we evaluate three learning paradigms: (i) supervised learning on expert over-read labels, (ii) Self-training that extends expert supervision to public ECGs, and (iii) multimodal contrastive learning with CLIP and NegCLIP. Across all settings, models trained with expert over-read data consistently outperform those trained on machine-read labels, especially for rare but clinically important conditions. Self-training and NegCLIP further demonstrate scalable strategies to propagate expert knowledge beyond labeled datasets. These findings highlight the essential role of expert over-reads in developing trustworthy and clinically aligned ECG AI systems.
APA
Kwak, S., Liu, F., Nizam, N.B., Richter, I., Uriel, N., Okin, P.M. & Sabuncu, M.R.. (2026). Beyond Machine Interpretation: Learning from Expert Over-Reads Improves ECG Diagnosis. Proceedings of The 9th International Conference on Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 315:36-55 Available from https://proceedings.mlr.press/v315/kwak26a.html.

Related Material