Key result
Machine learning using ECG and clinical features accurately predicts LV diastolic dysfunction with ~0.94 AUC.
Why the study?
LV diastolic dysfunction plays a major role in heart failure pathophysiology, but clinical tools to identify it before echocardiography remain imprecise.
Does a machine-learning model using clinical and ECG features accurately estimate myocardial relaxation and detect LV diastolic dysfunction compared to echocardiography?
Observational (n=1,202)
Yes
Does a machine-learning model using clinical and ECG features accurately estimate myocardial relaxation and detect LV diastolic dysfunction compared to echocardiography?
Effect estimate: AUC 0.88 and 0.94
A machine-learning model utilizing ECG and clinical features can accurately estimate myocardial relaxation and detect LV diastolic dysfunction, providing a potential cost-effective screening tool.
No takes yet. Share an insight, caveat, or question.
Captured external expert commentary on this paper, strongest first. Original sources are linked where available.
“The echo is not a difficult test. It's the most proven usable tool that we have in cardiology because it's easy to reproduce, low cost, and noninvasive – so we have all that we want in medicine.”
“The study focused on developing a quantitative estimation of a key echocardiographic measure. The results demonstrate a potentially significant new role for electrocardiography in cardiac testing and reducing overall healthcare costs.”
“Although we are excited about the prospects of such developments, we hold out for better evidence for their actual use.”
Supports noninvasive diastolic assessment from routine data; leaves open prospective validation before clinical adoption.
Kagiyama et al. (2020) conducted an observational in Left ventricular diastolic dysfunction (n=1,202). Machine-learning models using clinical and ECG features vs. Echocardiography was evaluated on Prediction of LV diastolic dysfunction based on multiple age- and sex-adjusted reference limits (AUC 0.88 and 0.94). A machine-learning model using clinical and ECG features accurately predicted left ventricular diastolic dysfunction with an AUC of 0.88 in the internal test set and 0.94 in the external test set.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: