{"id":"W7106852203","doi":"10.48448/v3zh-ae64","title":"Not What the Doctor Ordered: Surveying LLM-based De-identification and Quantifying Clinical Information Loss","year":2025,"lang":"","type":"other","venue":"Open MIND","topic":"","field":"","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Research Canada; Arthritis Research Centre of Canada; University of Alberta","funders":"","keywords":"Metric (unit); Set (abstract data type); Key (lock); Health care; Measure (data warehouse); MEDLINE; Risk assessment","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":[0.1617485,0.0009320022,0.001211492,0.01207701,0.001243154,0.006479037,0.003417945,0.001925491,0.001922923],"category_scores_gemma":[0.5577039,0.0004654868,0.001531209,0.01032293,0.00214993,0.006698433,0.004405508,0.002442622,0.0009441148],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.004384309,"about_ca_system_score_gemma":0.005212577,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006718361,"about_ca_topic_score_gemma":0.005161521,"domain_scores_codex":[0.8307612,0.10716,0.02404394,0.00924086,0.02737792,0.001416122],"domain_scores_gemma":[0.2692865,0.5920287,0.05420713,0.0359872,0.04684594,0.001644525],"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":[0.001462434,0.0003177687,0.3456245,0.006225432,0.001364376,0.0003240667,0.006769944,0.01410319,0.001702938,0.01601058,0.02866362,0.5774311],"study_design_scores_gemma":[0.0005367088,0.002145243,0.2974785,0.01769204,0.002638493,0.003546884,0.01590562,0.3046706,0.03952749,0.1161442,0.1987227,0.0009915393],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.43626,0.05612873,0.3808553,0.05727509,0.001705116,0.002451311,0.0302775,0.003914667,0.03113223],"genre_scores_gemma":[0.8018327,0.007491956,0.1645699,0.006774772,0.0006771017,0.001320845,0.01506324,0.0004921315,0.001777353],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.1617485,"threshold_uncertainty_score":0.855418,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1694193313871125,"score_gpt":0.4336797111412924,"score_spread":0.2642603797541799,"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."}}