{"id":"W4200044765","doi":"10.1016/j.cardfail.2021.12.004","title":"Predicting 30-Day Readmissions in Patients With Heart Failure Using Administrative Data: A Machine Learning Approach","year":2021,"lang":"en","type":"article","venue":"Journal of Cardiac Failure","topic":"Heart Failure Treatment and Management","field":"Medicine","cited_by":44,"is_retracted":false,"has_abstract":false,"ca_institutions":"Boehringer Ingelheim (Canada); University of Alberta","funders":"Mitacs","keywords":"Medicine; Receiver operating characteristic; Likelihood ratios in diagnostic testing; Machine learning; Decision tree; Predictive value; Predictive modelling; Artificial intelligence; Classifier (UML); Heart failure; Statistics; Artificial neural network; Emergency medicine; Internal medicine; Computer science; Mathematics","routes":{"ca_aff":true,"ca_fund":true,"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":[0.0007568033,0.0002641941,0.0008086779,0.0002194502,0.0001351127,0.00006955057,0.0001266396,0.000122876,0.00009059163],"category_scores_gemma":[0.0003997085,0.0001881523,0.0001911306,0.0004743078,0.00004385398,0.0004504281,0.0001506993,0.0008741019,0.000003978656],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001708598,"about_ca_system_score_gemma":0.0006308357,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001568634,"about_ca_topic_score_gemma":0.00002336656,"domain_scores_codex":[0.9976107,0.0003485801,0.0005759753,0.0003591644,0.0007561633,0.0003494025],"domain_scores_gemma":[0.9983973,0.0001030774,0.0003099584,0.0004540774,0.0004125537,0.0003230183],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0002072975,0.0006184212,0.9875405,0.0001117629,0.0007304297,0.0002361335,0.0005697559,0.0003807181,0.0004377287,0.00001559597,0.008456396,0.0006952239],"study_design_scores_gemma":[0.01098324,0.002491582,0.4203433,0.00232769,0.002517718,0.0004053437,0.007109951,0.002434083,0.0007418631,0.00001178086,0.5499514,0.0006819705],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9878004,0.0008788898,0.0006510331,0.00859121,0.000229567,0.0008106604,0.00006950246,0.00004403762,0.0009246922],"genre_scores_gemma":[0.9398601,0.0000300602,0.05904371,0.0001099259,0.0003468237,0.000004957355,0.0002715749,0.00004423878,0.0002886205],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5671972,"threshold_uncertainty_score":0.7672629,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03565351391030326,"score_gpt":0.300094118272528,"score_spread":0.2644406043622247,"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."}}