{"id":"W4317750115","doi":"10.1007/s41666-023-00125-6","title":"Can Patients with Dementia Be Identified in Primary Care Electronic Medical Records Using Natural Language Processing?","year":2023,"lang":"en","type":"article","venue":"Journal of Healthcare Informatics Research","topic":"Dementia and Cognitive Impairment Research","field":"Medicine","cited_by":18,"is_retracted":false,"has_abstract":false,"ca_institutions":"Women's College Hospital; University of Toronto; University Health Network; Health Sciences Centre; Toronto Rehabilitation Institute; Sunnybrook Health Science Centre; Public Health Ontario","funders":"Alzheimer Society Research Program; Alzheimer Society; Canadian Institutes of Health Research; Government of Ontario","keywords":"Dementia; Medical record; Cognition; Cohort; Medicine; Artificial intelligence; MEDLINE; Psychology; Natural language processing; Machine learning; Psychiatry; Computer science; Disease; Internal medicine","routes":{"ca_aff":true,"ca_fund":true,"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":[0.003059832,0.0002801084,0.0005974339,0.003379827,0.0006149735,0.002569909,0.0006513626,0.001373186,0.002100604],"category_scores_gemma":[0.04648427,0.0002335791,0.0007678106,0.002496261,0.0004487256,0.003547272,0.0008064574,0.0008289851,0.0006928777],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007719834,"about_ca_system_score_gemma":0.00158947,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.009971036,"about_ca_topic_score_gemma":0.01480589,"domain_scores_codex":[0.9969868,0.001134934,0.0007324597,0.0004436919,0.0004707103,0.0002313372],"domain_scores_gemma":[0.9806417,0.01284688,0.003574609,0.0005457465,0.002059342,0.0003317135],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0005480181,0.0004496597,0.9046497,0.001021005,0.0002166238,0.001426463,0.002179985,0.0002753709,0.001158265,0.0009055969,0.009770411,0.07739887],"study_design_scores_gemma":[0.0001870676,0.0005039533,0.9396605,0.001481181,0.0007390712,0.005041381,0.01391196,0.009694783,0.001966683,0.008172098,0.01851175,0.0001295672],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9492617,0.005675139,0.004501722,0.02013205,0.0004751285,0.0002688295,0.008688065,0.0001644382,0.01083302],"genre_scores_gemma":[0.9840439,0.001749866,0.005622425,0.003563258,0.0002832304,0.00012308,0.004004134,0.00001142647,0.0005987737],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.009971036,"threshold_uncertainty_score":0.01982599,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03382600970431461,"score_gpt":0.4133786813594748,"score_spread":0.3795526716551602,"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."}}