{"id":"W4417515038","doi":"10.1038/s41598-025-31562-5","title":"Predicting and classifying type 2 diabetes using a transparent ensemble model combining random forest, k-nearest neighbor, and neural networks","year":2025,"lang":"en","type":"article","venue":"Scientific Reports","topic":"Artificial Intelligence in Healthcare","field":"Health Professions","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"","keywords":"Interpretability; Random forest; Ensemble learning; Artificial neural network; Feature selection; Missing data; Decision tree; Leverage (statistics); Deep learning; Data pre-processing","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":["sts"],"consensus_categories":[],"category_scores_codex":[0.002413184,0.0002165827,0.0004169469,0.0002151997,0.003018465,0.0001669825,0.0001050262,0.0002309904,0.00001509849],"category_scores_gemma":[0.0005578048,0.000200386,0.00005092353,0.0005796553,0.0003413398,0.0002751722,0.0001515823,0.000666019,9.575194e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001125266,"about_ca_system_score_gemma":0.0003889683,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0005032478,"about_ca_topic_score_gemma":0.001589455,"domain_scores_codex":[0.996613,0.00029884,0.001124966,0.0008320164,0.0003103986,0.0008207624],"domain_scores_gemma":[0.99785,0.0006687178,0.0004057493,0.0005030764,0.0003432253,0.0002292331],"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":[0.00005025342,0.0000172153,0.9319198,0.0003221378,0.00001696512,0.00002983426,0.003006411,0.05990629,0.001402994,0.0001848374,0.000341787,0.002801488],"study_design_scores_gemma":[0.0002081701,0.00002266207,0.003520756,0.0008068706,0.00004864253,0.000007553582,0.001312497,0.984974,0.0001820008,0.008629467,0.0001304449,0.0001568798],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9760394,0.001930409,0.01222403,0.0003083202,0.007930728,0.001014716,0.000002334599,0.0001304911,0.0004195863],"genre_scores_gemma":[0.9985508,0.00004409874,0.000561056,0.0002111871,0.0001093284,0.0000520471,0.00001845023,0.00002720833,0.0004258707],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.928399,"threshold_uncertainty_score":0.9982795,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1487885684133083,"score_gpt":0.4222256248179109,"score_spread":0.2734370564046026,"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."}}