{"id":"W4404764897","doi":"10.1186/s12911-024-02776-w","title":"Development and validation of a rheumatoid arthritis case definition: a machine learning approach using data from primary care electronic medical records","year":2024,"lang":"en","type":"article","venue":"BMC Medical Informatics and Decision Making","topic":"Rheumatoid Arthritis Research and Therapies","field":"Medicine","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto; Sunnybrook Health Science Centre; College of Physicians and Surgeons of Ontario; University of Alberta; Simon Fraser University; University of Calgary","funders":"Mitacs; Arthritis Society","keywords":"Random forest; Medicine; Machine learning; Artificial intelligence; Decision tree; Feature (linguistics); Medical record; Data mining; Cluster analysis; Computer science; Health informatics; Gradient boosting; Support vector machine; Cross-validation; Public health; Internal medicine; Pathology","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.04528573,0.0007839324,0.0008695141,0.00546223,0.001181075,0.002587635,0.002626528,0.00123171,0.0007334666],"category_scores_gemma":[0.1100165,0.0003673102,0.001334397,0.002837914,0.001239281,0.001291074,0.00202098,0.001116518,0.0002443397],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.003557778,"about_ca_system_score_gemma":0.005712515,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01750765,"about_ca_topic_score_gemma":0.01851225,"domain_scores_codex":[0.9628094,0.02126992,0.004892016,0.004023907,0.006265917,0.0007388077],"domain_scores_gemma":[0.9031168,0.0607859,0.00948333,0.006726852,0.01904006,0.0008469372],"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":[0.000830148,0.001264833,0.8341165,0.0008968425,0.0005134595,0.0007874104,0.002510065,0.02595316,0.002810261,0.002448556,0.00407044,0.1237984],"study_design_scores_gemma":[0.0007264953,0.002365728,0.5074262,0.001118059,0.0007237255,0.002258185,0.003988094,0.4498088,0.01441229,0.004716629,0.01226463,0.0001911475],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8257803,0.000585285,0.1537302,0.0006559437,0.00008425894,0.01123313,0.004883486,0.0004337711,0.002613473],"genre_scores_gemma":[0.7380674,0.0001238163,0.2496552,0.0002177396,0.00003756544,0.004381787,0.007281983,0.00003103125,0.0002036078],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.04528573,"threshold_uncertainty_score":0.2394967,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06798890512528732,"score_gpt":0.3331619303872496,"score_spread":0.2651730252619622,"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."}}