{"id":"W4414481098","doi":"10.2196/79653","title":"Equivalence of Type 2 Diabetes Prevalence Estimates: Comparative Study of Similar Phenotyping Algorithms Using Electronic Health Record Data","year":2025,"lang":"en","type":"article","venue":"JMIR Public Health and Surveillance","topic":"Diabetes, Cardiovascular Risks, and Lipoproteins","field":"Medicine","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Center for Chronic Disease Prevention and Health Promotion; National Institute of Diabetes and Digestive and Kidney Diseases","keywords":"Electronic health record; Type 2 diabetes; Equivalence (formal languages); Diagnosis code; Public health; Health records; Harmonization","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.1665814,0.0006448805,0.001208436,0.003608952,0.0005305609,0.00388315,0.001925473,0.001704289,0.0016202],"category_scores_gemma":[0.4662242,0.0006228366,0.003031512,0.002743697,0.001794449,0.005097989,0.00359746,0.001515659,0.0004034644],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001564383,"about_ca_system_score_gemma":0.001206658,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001619606,"about_ca_topic_score_gemma":0.001110104,"domain_scores_codex":[0.764722,0.1925462,0.01179264,0.01137564,0.01796491,0.001598573],"domain_scores_gemma":[0.5230427,0.4066049,0.02694818,0.02112885,0.0206581,0.001617362],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.01603279,0.001227368,0.8142481,0.001103378,0.01256302,0.000136575,0.003648685,0.01011614,0.0008832179,0.004227205,0.001617973,0.1341955],"study_design_scores_gemma":[0.002801897,0.01512386,0.8943124,0.000779631,0.004181466,0.0004953063,0.003552521,0.06432401,0.002097915,0.007656479,0.004471427,0.0002031829],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9516324,0.002199824,0.037953,0.000607944,0.0002510145,0.001418725,0.001566877,0.0001623215,0.004207964],"genre_scores_gemma":[0.9851821,0.0003455345,0.01140646,0.0002865792,0.00007998938,0.0005835925,0.001873731,0.00006380223,0.0001782059],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.1665814,"threshold_uncertainty_score":0.8809772,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09914457068269433,"score_gpt":0.3887612442120023,"score_spread":0.289616673529308,"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."}}