{"id":"W4409323950","doi":"10.1093/ehjdh/ztaf034","title":"Development and multinational validation of an ensemble deep learning algorithm for detecting and predicting structural heart disease using noisy single-lead electrocardiograms","year":2025,"lang":"en","type":"article","venue":"European Heart Journal - Digital Health","topic":"ECG Monitoring and Analysis","field":"Medicine","cited_by":21,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Institutes of Health; Yale University; National Heart, Lung, and Blood Institute; Bristol-Myers Squibb Canada; National Institute on Aging; Bristol-Myers Squibb; Doris Duke Charitable Foundation","keywords":"Computer science; Multinational corporation; Artificial intelligence; Ensemble learning; Lead (geology); Algorithm; Cross-validation; Machine learning; Pattern recognition (psychology); Finance","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.005788517,0.001024937,0.0006056334,0.0008191644,0.0003230275,0.0004783369,0.001089903,0.0009164062,0.000823778],"category_scores_gemma":[0.006943753,0.0002677594,0.0005677766,0.0004387345,0.0003161387,0.0005171901,0.001048096,0.0009151812,0.000308774],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000623072,"about_ca_system_score_gemma":0.0009065178,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007466775,"about_ca_topic_score_gemma":0.006426233,"domain_scores_codex":[0.9990212,0.0004257819,0.00007840253,0.0002468482,0.0001461264,0.000081759],"domain_scores_gemma":[0.9970375,0.001315,0.0001911414,0.0003700909,0.0009403106,0.0001459701],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0008443301,0.001445545,0.1506829,0.0000773112,0.0007015728,0.0001869313,0.0001207381,0.521156,0.009844348,0.0005858823,0.004808112,0.3095464],"study_design_scores_gemma":[0.00005332828,0.0002242906,0.007963301,0.00000905318,0.00003953199,0.00003339933,0.00001968391,0.9886748,0.002532935,0.0002038397,0.0002371363,0.000008784263],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9299641,0.0003597728,0.06631222,0.000272638,0.0001031875,0.0001931594,0.0006565093,0.0008551748,0.001283184],"genre_scores_gemma":[0.9580749,0.00007668217,0.03898166,0.0001359669,0.00002094851,0.000138366,0.00155272,0.00002946698,0.0009893286],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.007466775,"threshold_uncertainty_score":0.03061301,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03511543079916488,"score_gpt":0.3273124225099162,"score_spread":0.2921969917107513,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}