{"id":"W4401813527","doi":"10.21203/rs.3.rs-4655592/v1","title":"Fusion of automatically learned rhythm and morphology features matches diagnostic criteria and enhances AI explainability","year":2024,"lang":"en","type":"preprint","venue":"Research Square","topic":"ECG Monitoring and Analysis","field":"Medicine","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"Abbott (Canada)","funders":"","keywords":"Interpretability; Rhythm; Term (time); Artificial intelligence; Causality (physics); Representation (politics); Computer science; Morphology (biology); Pattern recognition (psychology); Machine learning; Internal medicine; Medicine; Physics; Biology","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001673312,0.0002285811,0.0007323529,0.0004423166,0.000117497,0.0001122746,0.000110123,0.0004002458,0.00011676],"category_scores_gemma":[0.004024561,0.0001763222,0.0001250759,0.0002688424,0.0006097347,0.00002963931,0.001068257,0.001641051,0.00001065905],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000734183,"about_ca_system_score_gemma":0.0002213003,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0005282731,"about_ca_topic_score_gemma":0.00002926763,"domain_scores_codex":[0.9973881,0.0004739361,0.0003706308,0.0007152975,0.0006588117,0.0003932716],"domain_scores_gemma":[0.9968262,0.001829924,0.00006175096,0.0005619177,0.0004911489,0.0002290609],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.001986705,0.002078654,0.2067802,0.2680535,0.002449558,0.005321799,0.02328816,0.00005350974,0.1905475,0.001879964,0.0174061,0.2801543],"study_design_scores_gemma":[0.001720004,0.003196256,0.7704843,0.03050589,0.001179273,0.0005447601,0.007816368,0.006436842,0.05233164,0.1239557,0.0008431372,0.0009857768],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9787681,0.01223193,0.00004232967,0.007916583,0.0001429793,0.000533085,0.00004407712,0.0000811499,0.0002397381],"genre_scores_gemma":[0.9914711,0.006317007,0.0008785946,0.00001918022,0.0003154659,0.0001297374,0.00004410352,0.00002913256,0.0007956374],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5637041,"threshold_uncertainty_score":0.7190209,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0472262170924327,"score_gpt":0.4393618925189006,"score_spread":0.3921356754264679,"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."}}