{"id":"W2954286747","doi":"10.1371/journal.pone.0218760","title":"Feature selection and transformation by machine learning reduce variable numbers and improve prediction for heart failure readmission or death","year":2019,"lang":"en","type":"article","venue":"PLoS ONE","topic":"Heart Failure Treatment and Management","field":"Medicine","cited_by":74,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Ottawa","funders":"Department of Health, Government of Western Australia","keywords":"Feature selection; Receiver operating characteristic; Predictive modelling; Medicine; Multilayer perceptron; Heart failure; Machine learning; Artificial intelligence; Perceptron; Statistics; Computer science; Artificial neural network; Internal medicine; Mathematics","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003949633,0.000761369,0.0007805474,0.001272146,0.0002287617,0.0007720786,0.0005752792,0.0003311003,0.001792695],"category_scores_gemma":[0.01471492,0.000151244,0.001124453,0.00155691,0.0004205628,0.0006803676,0.0005777466,0.0008157421,0.0004902586],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003935713,"about_ca_system_score_gemma":0.000986849,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001828347,"about_ca_topic_score_gemma":0.001649551,"domain_scores_codex":[0.9985391,0.0008039616,0.0001489693,0.0002089241,0.0002193071,0.00007984069],"domain_scores_gemma":[0.9920025,0.006063486,0.0007291774,0.0005644908,0.0005292942,0.0001111248],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0009891408,0.001028653,0.4247286,0.0001875209,0.0006872189,0.0003899597,0.0002601427,0.07143307,0.01023971,0.0007184119,0.003468556,0.485869],"study_design_scores_gemma":[0.00024455,0.001418665,0.2744839,0.00007108161,0.0004183886,0.0005865299,0.0001929694,0.6996666,0.013546,0.006534203,0.00277643,0.00006063176],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"methods","genre_scores_codex":[0.8125929,0.0004828074,0.1830602,0.0007766303,0.0001056299,0.0001839155,0.0007246668,0.001041957,0.001031352],"genre_scores_gemma":[0.9371296,0.0001578315,0.06115815,0.00009141662,0.0000638499,0.0001070125,0.0008368458,0.00004883921,0.0004063738],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.003949633,"threshold_uncertainty_score":0.02088785,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0153443060725264,"score_gpt":0.2364064399268496,"score_spread":0.2210621338543232,"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."}}