{"id":"W4409413640","doi":"10.2147/clep.s504259","title":"Extracting Cognitive Impairment Assessment Information From Unstructured Notes in Electronic Health Records Using Natural Language Processing Tools: Validation with Clinical Assessment Data","year":2025,"lang":"en","type":"article","venue":"Clinical Epidemiology","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"U.S. National Library of Medicine; National Institutes of Health","keywords":"Health records; Medicine; Cognitive impairment; Cognition; Unstructured data; Natural language processing; Electronic health record; Data science; Information retrieval; Data mining; Computer science; Artificial intelligence; Psychiatry; Big data; Health care","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaresearch","research_integrity"],"consensus_categories":[],"category_scores_codex":[0.02162992,0.000289704,0.001220336,0.0002091417,0.0002477751,0.0001301585,0.001082252,0.0003435559,0.00001104805],"category_scores_gemma":[0.01940398,0.0002387789,0.0001047118,0.0005845072,0.0001588777,0.001651546,0.0007576743,0.002880854,0.000003417604],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006255402,"about_ca_system_score_gemma":0.003811759,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002470497,"about_ca_topic_score_gemma":0.0007798228,"domain_scores_codex":[0.9857318,0.008201541,0.003753814,0.001170217,0.0002769844,0.0008656852],"domain_scores_gemma":[0.9705215,0.02577445,0.002278674,0.001032193,0.0002180437,0.0001751931],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00006213968,0.00007483072,0.4855453,0.00005715459,0.00003579025,0.000003156452,0.0001394183,0.0006020773,8.106539e-7,0.0006539293,0.00003255842,0.5127928],"study_design_scores_gemma":[0.0009235165,0.0003158025,0.4707378,0.0003565318,0.00001581036,0.000005952317,0.0001890605,0.5260031,9.74509e-7,0.001172104,0.0001404349,0.0001388863],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3067439,0.0006093566,0.6813905,0.009477027,0.0009249192,0.0006610686,0.00002319873,0.0001226345,0.0000473956],"genre_scores_gemma":[0.7323129,0.0001166973,0.2603561,0.006329889,0.0001622309,0.00002117321,0.0006901589,0.00000903092,0.000001761692],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.5254011,"threshold_uncertainty_score":0.9994196,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1923096115347233,"score_gpt":0.5624915034290198,"score_spread":0.3701818918942965,"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."}}