Key result
Queen of Hearts AI support increased physician diagnostic discrimination (MRMC AUC) for post-ROSC occlusive myocardial infarction from 0.882 to 0.912 (Δ 0.030, 95% CI 0.008-0.058; p=0.018).
Why the study?
Interpreting post-ROSC ECGs to identify occlusive myocardial infarction requiring urgent reperfusion is challenging. Although AI-based decision-support tools show promise, their impact on physician decision-making remained unclear.
Does artificial intelligence-assisted ECG interpretation using the Queen of Hearts neural network improve physician diagnostic discrimination and accuracy for identifying occlusive myocardial infarction in post-cardiac arrest (post-ROSC) ECGs compared to unassisted physician interpretation?
Does artificial intelligence-assisted ECG interpretation using the Queen of Hearts neural network improve physician diagnostic discrimination and accuracy for identifying occlusive myocardial infarction in post-cardiac arrest (post-ROSC) ECGs compared to unassisted physician interpretation?
Answers are grounded in this paper and cite their sources.
AI-assisted ECG interpretation using the Queen of Hearts neural network significantly improved overall physician diagnostic discrimination (AUC 0.882 to 0.912) and specificity (84.64% to 90.43%) for post-cardiac arrest occlusive myocardial infarction, though it did not improve sensitivity for non-STEMI high-risk OMI and led to diagnostic errors when the algorithm was incorrect.
Interpretation of post–return of spontaneous circulation (ROSC) electrocardiograms (ECGs) is challenging, yet critical for identifying occlusive myocardial infarction (OMI) requiring urgent reperfusion. While artificial intelligence (AI)–based decision-support tools show promising results, their impact on physician decision-making remains unclear. In this multi-reader, multi-case (MRMC) study, 69 real-world post-ROSC ECGs were evaluated by 15 physicians from five groups at two timepoints (T1 [without AI] and T2 [with Queen of Hearts (QoH) neural network support]), yielding 2,070 interpretations. At T1, sensitivity/specificity were 77.97%/84.64% (LR + 5.08, LR − 0.26); at T2, 78.55%/90.43% (LR + 8.21, LR − 0.24). For the primary endpoint, MRMC AUC increased from 0.882 to 0.912 (Δ 0.030, 95% CI 0.008–0.058; p =0.018). The error rate fell by 2.32% (95% CI − 0.9 to + 5.6; p =0.156); odds of correct identification were numerically higher at T2 (aOR 1.29, 95% CI 0.98–1.69). Of 66 changed decisions, 64 (97%) followed QoH: the error rate fell by 5.03% where QoH was correct and rose by 10.56% where it was not. Sensitivity for non-STEMI high-risk OMI did not improve. Confidence increased, and within-group agreement rose through convergence on the algorithm output. The benefit of AI support is tied to algorithm accuracy.
No takes yet. Share an insight, caveat, or question.
Silwanis et al. (2026) studied Occlusive myocardial infarction (OMI) following cardiac arrest (post-ROSC) (n=69). Queen of Hearts (QoH) AI-assisted ECG interpretation vs. Standard unassisted physician ECG interpretation was evaluated on Reader-averaged empirical area under the ROC curve (MRMC AUC) for identifying occlusive myocardial infarction (MD 0.030, 95% CI 0.008-0.058, p=0.018). Queen of Hearts AI support increased physician diagnostic discrimination (MRMC AUC) for post-ROSC occlusive myocardial infarction from 0.882 to 0.912 (Δ 0.030, 95% CI 0.008-0.058; p=0.018).
Synapse has enriched 10 related papers. Consider them for comparative context: