{"id":"W4225166168","doi":"10.1101/2022.04.24.22274125","title":"Towards Equitable Patient Subgroup Performance by Gene-Expression-Based Diagnostic Classifiers of Acute Infection","year":2022,"lang":"en","type":"preprint","venue":"medRxiv","topic":"COVID-19 diagnosis using AI","field":"Medicine","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Institute of Infection and Immunity","funders":"Imperial College London; National Institute of Arthritis and Musculoskeletal and Skin Diseases; Università ta' Malta; Universitetet i Oslo; Sidra Medicine; National Institutes of Health; Cincinnati Children's Hospital Medical Center","keywords":"Machine learning; Artificial intelligence; Confounding; Classifier (UML); Leverage (statistics); Procalcitonin; Medicine; Computer science; Bioinformatics; Biology; Internal medicine; Sepsis","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":[0.02531865,0.00062805,0.001378681,0.002612702,0.0004895555,0.003283951,0.0008949729,0.00112174,0.002228736],"category_scores_gemma":[0.0768832,0.0002551701,0.001377394,0.001714969,0.0007367142,0.001299829,0.001688897,0.001758387,0.0009682744],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006838034,"about_ca_system_score_gemma":0.001125551,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001475253,"about_ca_topic_score_gemma":0.00137767,"domain_scores_codex":[0.988077,0.007231148,0.000983302,0.001697637,0.001585726,0.00042509],"domain_scores_gemma":[0.940815,0.03615554,0.007508091,0.01036077,0.004329375,0.0008311265],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.003166273,0.000325705,0.8110908,0.0004908491,0.002054064,0.0001660499,0.0005010157,0.01852062,0.01012273,0.003606317,0.01488611,0.1350694],"study_design_scores_gemma":[0.0006341817,0.002777719,0.529634,0.0009111752,0.002321762,0.001342102,0.001956124,0.3081264,0.04076856,0.07335792,0.03792778,0.0002422646],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8683224,0.004700716,0.09262501,0.0068246,0.0004224209,0.0004178023,0.01905834,0.001391058,0.006237684],"genre_scores_gemma":[0.9635171,0.0003515244,0.02803748,0.0008253521,0.0001709167,0.0001439719,0.006272026,0.0001397961,0.0005417832],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.02531865,"threshold_uncertainty_score":0.1338995,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0252927728636511,"score_gpt":0.2993615025637518,"score_spread":0.2740687297001007,"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."}}