{"id":"W4392160170","doi":"10.7717/peerj-cs.1896","title":"Ensemble machine learning reveals key features for diabetes duration from electronic health records","year":2024,"lang":"en","type":"article","venue":"PeerJ Computer Science","topic":"Diabetes Management and Research","field":"Medicine","cited_by":8,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"Ministero dell'Università e della Ricerca; European Commission; Dipartimenti di Eccellenza","keywords":"Diabetes mellitus; Random forest; Context (archaeology); Type 2 diabetes; Medicine; Health records; Kidney disease; Disease; Retinopathy; Artificial intelligence; Diabetic retinopathy; Computer science; Machine learning; Internal medicine; Geography; Health care; Endocrinology","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":[0.002756994,0.000553059,0.0008333594,0.003307231,0.0002866209,0.0008455874,0.0004555075,0.0006457824,0.0007393443],"category_scores_gemma":[0.008906887,0.0001081653,0.0009690846,0.002480672,0.0001241959,0.0006902029,0.0004129742,0.0008106685,0.000421169],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003791286,"about_ca_system_score_gemma":0.0005271087,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00900151,"about_ca_topic_score_gemma":0.007731575,"domain_scores_codex":[0.998813,0.0003836078,0.0001328658,0.0002899188,0.0001992687,0.0001812425],"domain_scores_gemma":[0.9948336,0.003206135,0.0005501804,0.0005114165,0.0007173194,0.0001813508],"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":[0.000727906,0.0004835146,0.7820879,0.0001434159,0.0007513626,0.0004040262,0.0001497993,0.03611269,0.002548553,0.0004119734,0.007917536,0.1682613],"study_design_scores_gemma":[0.00002817855,0.0002745672,0.4743029,0.00008084403,0.0004350277,0.0004537031,0.00025542,0.5168742,0.002284359,0.001355467,0.003598649,0.00005662002],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9677321,0.002182566,0.02191955,0.0006814991,0.0001729184,0.00004293191,0.005601796,0.0004879885,0.001178525],"genre_scores_gemma":[0.9859262,0.000349177,0.006872549,0.00006303803,0.0001051638,0.00001868779,0.006335212,0.00001296066,0.0003169871],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.00900151,"threshold_uncertainty_score":0.01789826,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01751837860891434,"score_gpt":0.3138604138712033,"score_spread":0.296342035262289,"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."}}