{"id":"W4393299512","doi":"10.1101/2024.03.27.24304781","title":"Transport-based transfer learning on Electronic Health Records: Application to detection of treatment disparities","year":2024,"lang":"en","type":"preprint","venue":"medRxiv","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Provincial Health Services Authority; University of British Columbia","funders":"","keywords":"Transfer of learning; Generalization; Computer science; Machine learning; Artificial intelligence; Binary classification; Divergence (linguistics); Population; Binary number; Health records; Feature (linguistics); Space (punctuation); Supervised learning; Data mining; Health care; Artificial neural network; Mathematics; Support vector machine","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"],"consensus_categories":[],"category_scores_codex":[0.0007099528,0.0003606836,0.0005153048,0.0004254105,0.0001324234,0.0000487926,0.0005864707,0.0001952989,0.000006614724],"category_scores_gemma":[0.00002696442,0.0003389404,0.0002215493,0.0004478848,0.00002320785,0.00003247118,0.00007220931,0.001284438,0.00003609478],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001079961,"about_ca_system_score_gemma":0.0009103816,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004177063,"about_ca_topic_score_gemma":0.002471189,"domain_scores_codex":[0.9971212,0.0003861132,0.0005992941,0.00101901,0.0003697919,0.0005046018],"domain_scores_gemma":[0.9985982,0.0001400648,0.0001423706,0.0008693372,0.00007761527,0.0001723797],"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":[0.0001494731,0.0002377664,0.02172041,0.003788935,0.0001119423,0.000007068496,0.004174221,0.3426165,0.001395923,0.01355471,0.00001057178,0.6122324],"study_design_scores_gemma":[0.0009833759,0.01333456,0.06709956,0.002424133,0.0001156633,0.00001024886,0.00006585051,0.8493166,0.02157961,0.009178021,0.03433605,0.001556336],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4574796,0.0005572625,0.5322856,0.007395725,0.0005160405,0.001152726,0.00001487671,0.0005027361,0.00009548666],"genre_scores_gemma":[0.9973262,0.0001596209,0.001153938,0.0002962663,0.0001203919,0.0006853822,0.00004316109,0.00004975111,0.0001653194],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.6106761,"threshold_uncertainty_score":0.9999062,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01730530607449782,"score_gpt":0.2954770456638658,"score_spread":0.278171739589368,"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."}}