{"id":"W4310490319","doi":"10.3389/fpubh.2022.1070870","title":"An NLP tool for data extraction from electronic health records: COVID-19 mortalities and comorbidities","year":2022,"lang":"en","type":"article","venue":"Frontiers in Public Health","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Medicine; ARDS; Artificial intelligence; Machine learning; Cause of death; Intensive care medicine; Mortality rate; Computer science; Natural language processing; Medical emergency; Disease; Pathology; Internal medicine","routes":{"ca_aff":true,"ca_fund":false,"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":[0.004625024,0.001054639,0.0008107809,0.006453095,0.001119014,0.001979915,0.0009508763,0.0007102817,0.01147616],"category_scores_gemma":[0.01727782,0.0005570092,0.001294768,0.004991992,0.0005877113,0.002196379,0.003342974,0.001074757,0.005568272],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001059123,"about_ca_system_score_gemma":0.004965672,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003742541,"about_ca_topic_score_gemma":0.004756849,"domain_scores_codex":[0.9973751,0.0005946861,0.0008608695,0.0005551278,0.000521817,0.00009230581],"domain_scores_gemma":[0.9896912,0.006905853,0.00156647,0.0005426228,0.001051078,0.0002427963],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0009533453,0.0004949151,0.08790033,0.01117161,0.0006514404,0.008267265,0.006516458,0.008699925,0.03201909,0.01241418,0.2368034,0.594108],"study_design_scores_gemma":[0.0006448003,0.0004079712,0.16295,0.002250779,0.0006458767,0.009260704,0.004117315,0.1807273,0.05553048,0.0503394,0.5325842,0.0005411262],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04382491,0.0007206048,0.5474312,0.003635321,0.0002397124,0.004547531,0.2259447,0.1621509,0.01150508],"genre_scores_gemma":[0.09940392,0.0008238921,0.7572691,0.0009368028,0.0001444104,0.003828247,0.1297486,0.003104963,0.004740068],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01147616,"threshold_uncertainty_score":0.03839159,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09854911421712703,"score_gpt":0.3898702296378819,"score_spread":0.2913211154207548,"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."}}