{"id":"W4221120819","doi":"10.1016/j.compbiomed.2022.105449","title":"Handling missing values in machine learning to predict patient-specific risk of adverse cardiac events: Insights from REFINE SPECT registry","year":2022,"lang":"en","type":"article","venue":"Computers in Biology and Medicine","topic":"Acute Myocardial Infarction Research","field":"Medicine","cited_by":40,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Ottawa; Artificial Intelligence in Medicine (Canada); University of Calgary","funders":"National Center for Advancing Translational Sciences; National Heart, Lung, and Blood Institute; National Institutes of Health; Cedars-Sinai Medical Center; Dr. Miriam and Sheldon G. Adelson Medical Research Foundation; Sanofi; GE Healthcare; Astellas Pharma US","keywords":"Mace; Missing data; Imputation (statistics); Categorical variable; Receiver operating characteristic; Regression; Medicine; Regression analysis; Linear regression; Statistics; Mathematics; Internal medicine","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":[0.000659043,0.0001674673,0.0006875866,0.0005602041,0.0001482954,0.000001360202,0.0001269648,0.00008391382,0.00007304249],"category_scores_gemma":[0.000422385,0.0001394366,0.00006179952,0.000506593,0.0002360914,0.00003084783,0.0003074912,0.0009011946,0.00000165358],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001800251,"about_ca_system_score_gemma":0.00005330245,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001285796,"about_ca_topic_score_gemma":0.00001073085,"domain_scores_codex":[0.997847,0.0006856917,0.0005155017,0.0004502225,0.0002462832,0.0002553124],"domain_scores_gemma":[0.9988229,0.0005230343,0.0001635047,0.0002839189,0.00005100102,0.0001556455],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.001903952,0.0001659347,0.9229852,0.00005180841,0.0002767954,0.0002251915,0.00579409,0.001205649,0.0113578,0.0001032967,0.00271864,0.05321166],"study_design_scores_gemma":[0.0131999,0.007812657,0.7737932,0.001705161,0.0002143772,0.0001160127,0.006566212,0.009653223,0.001144757,0.002726123,0.1825163,0.000552017],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9850942,0.009512065,0.0009741021,0.002679575,0.0008832956,0.0004319677,0.00004332888,0.00003175738,0.0003496839],"genre_scores_gemma":[0.9939542,0.00245892,0.002142407,0.0007978517,0.0002968867,0.00002481121,0.0002684122,0.00001647739,0.00004000959],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.1797977,"threshold_uncertainty_score":0.5686061,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01944927597073007,"score_gpt":0.3050690462089197,"score_spread":0.2856197702381896,"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."}}