{"id":"W4311059376","doi":"10.1101/2022.11.29.22282888","title":"Tools for categorization of diagnostic codes in hospital data: Operationalizing CCSR into a patient data repository","year":2022,"lang":"en","type":"preprint","venue":"medRxiv","topic":"Medical Coding and Health Information","field":"Health Professions","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"Sinai Health System; St. Michael's Hospital; University of Toronto","funders":"Natural Sciences and Engineering Research Council of Canada; Canadian Institutes of Health Research; Alliance de recherche numérique du Canada; Canadian Frailty Network; University of Toronto; University Health Network","keywords":"Coding (social sciences); Operationalization; ICD-10; Categorization; Diagnosis code; Androstenediol; Medicine; Medical classification; Data mining; Computer science; Statistics; Artificial intelligence; Mathematics; Psychiatry; Environmental health; Nursing; Internal medicine","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":[0.03135261,0.001783358,0.001334951,0.02313755,0.00204829,0.007799477,0.003938311,0.0009146986,0.003871215],"category_scores_gemma":[0.1262263,0.001030463,0.001872345,0.01661763,0.001324242,0.004240542,0.006383531,0.001906244,0.003468715],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.006680904,"about_ca_system_score_gemma":0.01736608,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.08532761,"about_ca_topic_score_gemma":0.08176697,"domain_scores_codex":[0.9751248,0.006527252,0.004990657,0.005309817,0.007135233,0.0009121347],"domain_scores_gemma":[0.9165946,0.02974977,0.01145225,0.01468779,0.02456927,0.002946392],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0003970641,0.0002897808,0.1541382,0.001518609,0.0005010527,0.0006083316,0.00510711,0.01143224,0.005085269,0.0185429,0.1679505,0.634429],"study_design_scores_gemma":[0.0004084204,0.000358993,0.1736081,0.002984534,0.0004054581,0.001879266,0.006619314,0.4038399,0.03789927,0.04282872,0.3283438,0.0008241542],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.04569741,0.0006012986,0.7693666,0.005150786,0.0002624528,0.006698126,0.06311786,0.1001588,0.00894675],"genre_scores_gemma":[0.06963506,0.0001863343,0.8868359,0.0004288897,0.00005381275,0.001716943,0.03820682,0.001907199,0.001029084],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.08532761,"threshold_uncertainty_score":0.1696619,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3259460652604069,"score_gpt":0.472216701731843,"score_spread":0.1462706364714362,"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."}}