{"id":"W3127218785","doi":"10.1038/s41746-021-00394-8","title":"Predicting adverse outcomes due to diabetes complications with machine learning using administrative health data","year":2021,"lang":"en","type":"article","venue":"npj Digital Medicine","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":121,"is_retracted":false,"has_abstract":true,"ca_institutions":"Vector Institute; Lunenfeld-Tanenbaum Research Institute; Trillium Health Centre; Public Health Ontario; University of Toronto","funders":"Canadian Institutes of Health Research; Ontario Ministry of Health and Long-Term Care; Mitacs; Canada Research Chairs; Government of Canada","keywords":"Medicine; Health care; Decision tree; Population; Adverse effect; Diabetes mellitus; Machine learning; Hypoglycemia; Diabetes management; Emergency medicine; Artificial intelligence; Intensive care medicine; Computer science; Type 2 diabetes; Environmental health; Internal medicine","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0004677494,0.0002518403,0.0005006684,0.0001353024,0.0004085381,0.0001268633,0.001021969,0.00004170588,0.00002111662],"category_scores_gemma":[0.002122845,0.0001962137,0.0000280643,0.000869349,0.000100386,0.0008075115,0.0008625872,0.0005181677,0.00001850428],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001142719,"about_ca_system_score_gemma":0.0005347597,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000301729,"about_ca_topic_score_gemma":0.0002415461,"domain_scores_codex":[0.9973723,0.0001769864,0.000510211,0.0008486055,0.0005864059,0.0005055385],"domain_scores_gemma":[0.9969522,0.0006742954,0.0002797813,0.001374357,0.0002450783,0.000474242],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000006944252,0.00008530734,0.9658809,0.0001186657,0.00006135493,0.0001042589,0.002299196,0.001654602,0.0000451891,0.002389994,0.0001992644,0.02715427],"study_design_scores_gemma":[0.001493706,0.001927429,0.4834214,0.001577246,0.00003312874,0.0003437879,0.001807675,0.4946496,0.00004803378,0.0003309296,0.01371036,0.0006566406],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.4610583,0.0009494892,0.3792357,0.147423,0.0006516331,0.001141293,0.00032077,0.001095831,0.008124013],"genre_scores_gemma":[0.9646378,0.000003812339,0.03158727,0.003002943,0.0001236797,0.00001321679,0.0003997691,0.00002725529,0.000204267],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5035795,"threshold_uncertainty_score":0.8001361,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1133398611873404,"score_gpt":0.3866541926946153,"score_spread":0.2733143315072749,"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."}}