{"id":"W3191709143","doi":"10.1093/jamia/ocab128","title":"Improving domain adaptation in de-identification of electronic health records through self-training","year":2021,"lang":"en","type":"article","venue":"Journal of the American Medical Informatics Association","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"Google (Canada); University of Toronto","funders":"","keywords":"Computer science; Identification (biology); Software deployment; Domain (mathematical analysis); Domain adaptation; Artificial intelligence; Task (project management); Machine learning; Adaptation (eye); Test data; Data mining; Classifier (UML)","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":[],"consensus_categories":[],"category_scores_codex":[0.007176252,0.00007256525,0.0003074239,0.0001007881,0.00008567309,0.00005259673,0.0005580304,0.00005789829,0.000002467168],"category_scores_gemma":[0.004155719,0.00005970856,0.00009527148,0.0009137367,0.00002590979,0.0004975942,0.00009548193,0.0007412749,8.831182e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001614472,"about_ca_system_score_gemma":0.003451908,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003420127,"about_ca_topic_score_gemma":0.0001812542,"domain_scores_codex":[0.9962947,0.0007592287,0.00144674,0.00006354263,0.001129392,0.0003064451],"domain_scores_gemma":[0.9931545,0.0004887094,0.005695374,0.0002089431,0.0003684619,0.00008399071],"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":[0.00003144185,0.000290155,0.134398,0.00050353,0.0001974963,0.000009222146,0.2549462,0.01404718,0.0003541464,0.01585781,0.0006385449,0.5787263],"study_design_scores_gemma":[0.0006740621,0.0003213313,0.08474021,0.0002729638,0.00001405704,0.0001095896,0.01069725,0.8935282,0.0001994697,0.008673373,0.000642308,0.0001271939],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4294213,0.00008107493,0.5564542,0.01358155,0.0003068803,0.0000787087,7.854704e-7,0.00001509883,0.00006039359],"genre_scores_gemma":[0.8908061,0.0002227102,0.1069898,0.001880632,0.0000804251,0.000002113738,0.000001548826,0.000005520444,0.00001107876],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.879481,"threshold_uncertainty_score":0.6123536,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01171142378810457,"score_gpt":0.3038563359614966,"score_spread":0.292144912173392,"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."}}