{"id":"W2963824827","doi":"10.2196/14083","title":"Developing a Standardization Algorithm for Categorical Laboratory Tests for Clinical Big Data Research: Retrospective Study","year":2019,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Electronic Health Records Systems","field":"Health Professions","cited_by":18,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Standardization; Categorical variable; Computer science; Data mining; Algorithm; Data science; Medicine; Medical physics; Machine learning","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.03836693,0.001284153,0.001178138,0.007641841,0.001272474,0.003635244,0.002083923,0.001365609,0.001577642],"category_scores_gemma":[0.06684607,0.0008896972,0.002454191,0.007015485,0.001234478,0.003318967,0.002490605,0.00150864,0.000828578],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001594438,"about_ca_system_score_gemma":0.006691326,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003119034,"about_ca_topic_score_gemma":0.002359428,"domain_scores_codex":[0.9783996,0.009317754,0.004138049,0.003679214,0.004017147,0.0004483067],"domain_scores_gemma":[0.9573129,0.01710627,0.006264234,0.004269608,0.01423287,0.0008140174],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.0006219439,0.0004616617,0.3255992,0.001140407,0.0008995581,0.0004753095,0.001324888,0.0377186,0.005824716,0.01857306,0.01544312,0.5919175],"study_design_scores_gemma":[0.0004394489,0.000903743,0.07244167,0.0006396865,0.0008591156,0.001725602,0.001772835,0.8275323,0.02192734,0.03506273,0.03644304,0.0002524648],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.02136741,0.0003533557,0.9744151,0.0003741218,0.00008666526,0.00115641,0.0005435011,0.001107819,0.0005956562],"genre_scores_gemma":[0.08382454,0.000249779,0.9126066,0.0001428015,0.00005913058,0.001522768,0.001272586,0.0001121891,0.0002095456],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.03836693,"threshold_uncertainty_score":0.2029061,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.5084334879671147,"score_gpt":0.6169982422761427,"score_spread":0.108564754309028,"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."}}