{"id":"W4225114680","doi":"10.1101/2022.04.27.22274400","title":"Using Primary Care Text Data and Natural Language Processing to Monitor COVID-19 in Toronto, Canada","year":2022,"lang":"en","type":"preprint","venue":"medRxiv","topic":"COVID-19 diagnosis using AI","field":"Medicine","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Coronavirus disease 2019 (COVID-19); Context (archaeology); Medicine; Medical record; Artificial intelligence; Natural language processing; Pediatrics; Family medicine; Computer science; Geography; 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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0004560032,0.0003924928,0.0006849734,0.0001818309,0.0001777988,0.00008790437,0.0007170567,0.0001592369,0.0001832754],"category_scores_gemma":[0.00113301,0.0004080087,0.00004268388,0.0002467429,0.00005094855,0.0001235376,0.003612536,0.0009661813,6.989763e-7],"about_ca_system_candidate":true,"about_ca_system_consensus":true,"about_ca_system_score_codex":0.01398292,"about_ca_system_score_gemma":0.008966812,"about_ca_topic_candidate":true,"about_ca_topic_consensus":true,"about_ca_topic_score_codex":0.894107,"about_ca_topic_score_gemma":0.8786748,"domain_scores_codex":[0.9969704,0.0001502413,0.0004454262,0.00122416,0.0007603429,0.000449416],"domain_scores_gemma":[0.9975541,0.0002397331,0.0001775979,0.001476672,0.00008255104,0.0004693193],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0006247149,0.0003088728,0.7742029,0.03205092,0.0002230662,0.007340217,0.03782107,0.003182725,0.005908365,0.000002720107,0.01891343,0.119421],"study_design_scores_gemma":[0.00474594,0.0001805733,0.5789354,0.006600498,0.0009409502,0.0002852416,0.04235984,0.0227753,0.0005344524,0.000007444514,0.3397835,0.002850855],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9385579,0.05055013,0.00003978111,0.007711499,0.001102029,0.001192226,0.0005782021,0.0001116233,0.0001565949],"genre_scores_gemma":[0.9683134,0.0001010138,0.002658496,0.02696303,0.0004326665,0.00008472222,0.001219802,0.00009004738,0.0001368361],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.32087,"threshold_uncertainty_score":0.9998372,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04822252353629335,"score_gpt":0.3791056391960739,"score_spread":0.3308831156597805,"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."}}