Deep learning-based electrocardiographic differentiation of Takotsubo syndrome from acute anterior myocardial infarction: explainability, human-AI interaction, and a utilization bias hypothesis

Background Artificial intelligence (AI) may improve electrocardiographic differentiation of Takotsubo syndrome from acute anterior ST-segment elevation myocardial infarction (STEMI), but its impact on human diagnostic performance, the added value of visual explanations, and differences by reader background and experience remain uncertain. Methods Using 12-lead electrocardiogram (ECG) images from two hospitals, we developed and externally validated an ensemble of three deep learning models after self-supervised pretraining and fine-tuning. The diagnostic task used ECGs recorded within 14 days after symptom onset, regardless of reperfusion status. We conducted a four-phase reader study. Phases 1–3 compared unaided interpretation, a reference sheet, and additional AI assistance. In Phase 4, readers evaluated a new ECG set after randomization to reference-sheet support alone, with AI predictions, or with AI predictions plus gradient-weighted class activation mapping (Grad-CAM) heatmaps. Diagnostic correctness was analyzed using generalized linear mixed-effects models. Exploratory analyses examined differences by cardiologist status and experience. Results Development and external-validation cohorts, including healthy controls, comprised 832 and 95 participants, respectively. For Takotsubo–STEMI discrimination, external-validation accuracy was 0.914, with areas under the receiver operating characteristic and precision-recall curves of 0.970 and 0.984. Seventy healthcare providers completed the reader study. The estimated probability of correct diagnosis increased from 0.632 without assistance to 0.864 with AI assistance (odds ratio [OR], 3.71; P<0.001). In randomized Phase 4, AI assistance improved accuracy vs reference-sheet support alone (OR, 4.28; 95% CI, 3.08–5.95; P<0.001), and Grad-CAM provided additional benefit (OR, 1.53; 95% CI, 1.04–2.25; P=0.033). Exploratory analyses showed greater improvement among less experienced readers; with AI assistance, non-cardiologists outperformed cardiologists. AI-assisted readers remained less accurate overall than the model alone. Conclusions An ECG image-based ensemble differentiated Takotsubo syndrome from anterior STEMI and improved reader accuracy. Visual explanations provided additional benefit, while experience-related differences highlight the importance of effective human-AI interaction. Clinical Perspective What is new? An electrocardiogram (ECG) image-based ensemble deep learning model differentiated Takotsubo syndrome from acute anterior ST-segment elevation myocardial infarction with an external-validation accuracy of 0.914 and ROC AUC of 0.970 and exceeded mean reader accuracy on the reader-study ECG sets. In a human diagnostic test, displaying the model’s prediction improved readers’ diagnostic performance, and adding gradient-weighted class activation mapping (Grad-CAM) explanation produced further gains. Exploratory analyses showed greater improvement and higher AI-assisted accuracy among less experienced readers, with reversal of cardiologists’ unaided advantage under AI assistance, motivating a utilization bias hypothesis. What Are the Clinical Implications? The model exceeded mean unaided and AI-assisted reader accuracy on the reader-study ECG sets, supporting its potential to aid differentiation of Takotsubo syndrome from acute anterior myocardial infarction. AI assistance improved reader diagnostic accuracy, with further benefit from Grad-CAM visualizations, supporting further evaluation of these tools in clinical practice. Conventional clinical expertise may not ensure effective AI use, warranting further studies to test the utilization bias hypothesis and determine whether training in integrating AI predictions can improve performance across reader backgrounds and experience levels.