{"id":"W2989894618","doi":"10.1101/19012518","title":"Labelling chest x-ray reports using an open-source NLP and ML tool for text data binary classification","year":2019,"lang":"en","type":"preprint","venue":"medRxiv","topic":"COVID-19 diagnosis using AI","field":"Medicine","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"St. Michael's Hospital; University of Toronto","funders":"University of Toronto","keywords":"Artificial intelligence; Python (programming language); Medicine; Natural language processing; Computer science; Machine learning; Radiology; Generalizability theory; Chest radiograph; Radiography; Programming language","routes":{"ca_aff":true,"ca_fund":true,"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":[0.00316006,0.002169867,0.0009224495,0.005873396,0.0007289342,0.001953944,0.002371452,0.002126207,0.009258727],"category_scores_gemma":[0.01679861,0.0005644258,0.002373927,0.002411849,0.0006244843,0.002111842,0.002032025,0.002264441,0.01012764],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001464544,"about_ca_system_score_gemma":0.002147145,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007737595,"about_ca_topic_score_gemma":0.009151272,"domain_scores_codex":[0.997306,0.0005863442,0.0004207705,0.0009579971,0.0005959466,0.0001329325],"domain_scores_gemma":[0.9911634,0.005860098,0.0006043569,0.0008886615,0.001221713,0.0002618075],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001192831,0.001074107,0.01790748,0.003746884,0.0004757067,0.001860388,0.0008220415,0.05255995,0.01984758,0.004355115,0.2056843,0.6904737],"study_design_scores_gemma":[0.0002348105,0.0002863612,0.008986966,0.000323635,0.0001493135,0.0007732667,0.0003213582,0.8822907,0.0268893,0.01161903,0.0679932,0.0001320656],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05458613,0.002090112,0.4877539,0.001969356,0.0008690298,0.001889477,0.1424789,0.3035422,0.004820936],"genre_scores_gemma":[0.1614118,0.0005810506,0.6064962,0.000785941,0.0002644813,0.001875794,0.2211211,0.002869711,0.004593796],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.009258727,"threshold_uncertainty_score":0.03097349,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.204527997033697,"score_gpt":0.3973192601566911,"score_spread":0.1927912631229941,"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."}}