{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003369679,0.0003048225,0.0005934926,0.003838341,0.0006210497,0.00140739,0.0007197302,0.0007629745,0.001420045],"category_scores_gemma":[0.02848145,0.000431226,0.0008486026,0.00395727,0.0008046397,0.002373578,0.001366625,0.001122816,0.0003728905],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008460988,"about_ca_system_score_gemma":0.001253299,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005259994,"about_ca_topic_score_gemma":0.00516297,"domain_scores_codex":[0.9925932,0.001762966,0.002505522,0.001201224,0.00153297,0.0004041031],"domain_scores_gemma":[0.9593925,0.01518625,0.02097568,0.001391518,0.002228087,0.0008258861],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0001395692,0.0001163981,0.9945182,0.0001797159,0.00005359454,0.0002962585,0.0009910358,0.00003461185,0.0002475264,0.00003922702,0.0003110711,0.00307278],"study_design_scores_gemma":[0.00002870691,0.0002580278,0.9909794,0.0002424155,0.00008510419,0.002567128,0.003717728,0.000755091,0.0003908406,0.0001079195,0.0008360171,0.00003164974],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.997071,0.0005022855,0.0003503033,0.00006520219,0.000004317807,0.0001154181,0.001561577,0.000009878271,0.0003199677],"genre_scores_gemma":[0.9964471,0.0005064781,0.0008190529,0.0001081679,0.00001409225,0.0001295879,0.001865261,0.000009851456,0.000100365],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.005259994,"threshold_uncertainty_score":0.01782078,"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."}}