{"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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","sts"],"consensus_categories":[],"category_scores_codex":[0.006975028,0.0002343116,0.0005182026,0.0004837406,0.001327291,0.0003262993,0.001978502,0.00007091176,0.00005421902],"category_scores_gemma":[0.0005772695,0.0002766807,0.00003616183,0.0005772169,0.00007811224,0.001627731,0.0006071175,0.0009118973,6.975716e-7],"about_ca_system_candidate":true,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002641102,"about_ca_system_score_gemma":0.008785647,"about_ca_topic_candidate":true,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.03395084,"about_ca_topic_score_gemma":0.003003819,"domain_scores_codex":[0.9941241,0.002114612,0.000781375,0.001172857,0.0005519736,0.001255098],"domain_scores_gemma":[0.9966838,0.0003457619,0.0005075777,0.001737044,0.00004913619,0.000676698],"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":[0.00007281019,0.000324541,0.2011239,0.0006855301,0.00005881218,0.000009816104,0.01627635,0.0006873452,5.808289e-7,0.06619254,0.3378161,0.3767517],"study_design_scores_gemma":[0.0006566438,0.0008838829,0.01174637,0.00001098338,0.000001755384,0.00002657206,0.004221383,0.3306772,1.503214e-7,0.01340854,0.6380974,0.0002690659],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01578298,0.004165733,0.8586502,0.1160652,0.002825381,0.001255358,0.0008328634,0.0003892717,0.00003294114],"genre_scores_gemma":[0.6365713,0.002354713,0.3033265,0.05112249,0.0005429505,0.0008519258,0.004822845,0.00008728576,0.0003200757],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.6207883,"threshold_uncertainty_score":0.9999728,"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."}}