Key result
Multiclass and binary AI-ECG models show similar diagnostic accuracy for classifying LVEF.
Why the study?
The comparative performance of binary versus multiclass ECG-based deep learning models across reduced, mildly reduced, and preserved LVEF strata within a shared pipeline had not been established.
Does a multiclass deep learning ECG model improve LVEF classification accuracy compared to binary models across reduced, mildly reduced, and preserved LVEF strata?
Observational (n=8,226)
No
Does a multiclass deep learning ECG model improve LVEF classification accuracy compared to binary models across reduced, mildly reduced, and preserved LVEF strata?
Answers are grounded in this paper and cite their sources.
Binary and multiclass AI-ECG models perform comparably for LVEF classification, though both exhibit poor discrimination in the mildly reduced (41–49%) intermediate LVEF range.
LVEF is classified into reduced (≤40%), mildly reduced (41–49%), and preserved (≥50%) categories. ECG-based deep learning models may be formulated as binary or multiclass classifiers, but their comparative performance across all three strata in a shared pipeline has not been established. In the present study, we trained an autoencoder (AE) on 12-lead median-beat ECGs from 10,430 ECG-echocardiography pairs from 8,226 patients at the University Medical Centre Groningen to learn latent representations, which were subsequently classified by a multilayer perceptron (MLP) classifier. We systematically compared binary and multiclass LVEF classification across three clinically relevant LVEF thresholds: reduced (≤40%), mildly reduced (41–49%), and preserved (≥50%). Binary models achieved AUROCs of 0.897, 0.664, and 0.849 for ≤40%, 41–49%, and ≥50%, respectively, with comparable multiclass one-vs-rest AUROCs of 0.896, 0.660, and 0.845 (all P > 0.05). Both formulations performed poorest in the intermediate stratum. Our findings demonstrate that binary and multiclass AI–ECG LVEF models performed similarly across strata within the same AE–MLP framework, but both were weakest in the intermediate (41–49%) range.
No takes yet. Share an insight, caveat, or question.
Rama et al. (2026) conducted an observational in Left ventricular ejection fraction (LVEF) classification (n=8,226). Multiclass deep learning ECG-based LVEF classification (autoencoder and multilayer perceptron) vs. Binary deep learning ECG-based LVEF classification was evaluated on AUROC for LVEF classification across reduced (≤40%), mildly reduced (41–49%), and preserved (≥50%) categories (Multiclass AUROC 0.896 (≤40%), 0.660 (41–49%), 0.845 (≥50%) vs Binary AUROC 0.897, 0.664, 0.849, p=>0.05). Binary and multiclass AI-ECG models performed similarly for LVEF classification.
Synapse has enriched 10 related papers. Consider them for comparative context: