{"id":"W4229439926","doi":"10.2196/38482","title":"Prevalence of Sensitive Terms in Clinical Notes Using Natural Language Processing Techniques: Observational Study","year":2022,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Health Literacy and Information Accessibility","field":"Health Professions","cited_by":11,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Patient-Centered Outcomes Research Institute","keywords":"Preprint; Observational study; Confidentiality; Electronic health record; Computer science; Medicine; Internet privacy; Computer security; Health care; World Wide Web; Pathology; Political science","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.008699737,0.0001368977,0.0003880273,0.0001920082,0.0004817901,0.00001015994,0.0003605305,0.0001661222,0.0005665678],"category_scores_gemma":[0.00219857,0.0001116137,0.00005973797,0.0005505598,0.0001312174,0.001094018,0.000488673,0.002110572,0.000009367985],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000251246,"about_ca_system_score_gemma":0.001329859,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00007689367,"about_ca_topic_score_gemma":0.00001637315,"domain_scores_codex":[0.9926453,0.0009843615,0.004664049,0.000126049,0.001206712,0.0003735762],"domain_scores_gemma":[0.9966708,0.0011201,0.001504021,0.0002806215,0.0002569851,0.0001674395],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0001129821,0.0004863593,0.8251292,0.002922231,0.000007224312,0.000005810798,0.1145524,0.00001123715,0.000003907297,0.00004983304,0.0001658761,0.05655298],"study_design_scores_gemma":[0.001614303,0.0002725403,0.5878415,0.0009589501,0.00001290309,0.000009513682,0.09447381,0.3134374,0.00001653359,0.00003939818,0.001095707,0.0002274792],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.995599,0.00003125811,0.001075355,0.0001606181,0.0002943033,0.001946425,0.00005421625,0.0000806518,0.0007582259],"genre_scores_gemma":[0.9878729,0.000005386404,0.006525764,0.00511413,0.0001156215,0.000257315,0.00004739407,0.000008579809,0.00005292831],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3134262,"threshold_uncertainty_score":0.9169503,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1664574439244027,"score_gpt":0.5563529683678831,"score_spread":0.3898955244434803,"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."}}