{"id":"W4311827969","doi":"10.1371/journal.pdig.0000150","title":"Using Primary Care Clinical Text Data and Natural Language Processing to Identify Indicators of COVID-19 in Toronto, Canada","year":2022,"lang":"en","type":"article","venue":"PLOS Digital Health","topic":"Data-Driven Disease Surveillance","field":"Medicine","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"North York General Hospital; University of Toronto","funders":"","keywords":"Coronavirus disease 2019 (COVID-19); Medicine; Medical record; Retrospective cohort study; Severe acute respiratory syndrome; Primary care; Artificial intelligence; Family medicine; Pediatrics; Natural language processing; Computer science; Internal medicine; Disease","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0003411547,0.0001137631,0.0004275397,0.00008369164,0.0000956733,0.00002651977,0.0002792095,0.00002228623,0.00003157396],"category_scores_gemma":[0.0005366266,0.0001153522,0.0000209154,0.0002783261,0.0000537294,0.000299081,0.0007425533,0.0002317613,4.505014e-7],"about_ca_system_candidate":true,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002864865,"about_ca_system_score_gemma":0.008783129,"about_ca_topic_candidate":true,"about_ca_topic_consensus":true,"about_ca_topic_score_codex":0.2158582,"about_ca_topic_score_gemma":0.3733152,"domain_scores_codex":[0.998113,0.0001064035,0.000530197,0.0004348685,0.0005353577,0.0002801897],"domain_scores_gemma":[0.9985984,0.0001112089,0.000196621,0.0005035663,0.00002961695,0.0005605711],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.000351041,0.0003085358,0.8704051,0.003850606,0.00003536658,0.0002260329,0.00227813,0.000007918168,0.00002017667,0.000002894465,0.002286885,0.1202274],"study_design_scores_gemma":[0.002073461,0.0003107779,0.964819,0.0004249796,0.00003193316,0.00005827389,0.02130264,0.0008293738,0.000005298227,0.000002601779,0.009851104,0.0002906156],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.978673,0.01361909,0.000007992183,0.0003953336,0.0001017018,0.0005635464,0.00619818,0.00003432129,0.0004068078],"genre_scores_gemma":[0.9924419,0.00002028425,0.0002649641,0.004050177,0.00004970958,0.000007767706,0.003125353,0.00002109942,0.00001874843],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.1574569,"threshold_uncertainty_score":0.9968361,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05818390352170421,"score_gpt":0.4301188475247922,"score_spread":0.371934944003088,"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."}}