{"id":"W4410804643","doi":"10.1214/24-aoas2009","title":"A privacy-preserved and high-utility synthesis strategy for risk-based stratified subgroups of the Canadian scleroderma patient registry data","year":2025,"lang":"en","type":"article","venue":"The Annals of Applied Statistics","topic":"Privacy-Preserving Technologies in Data","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"McGill University; University of Alberta","funders":"","keywords":"Medicine; Data science; Family medicine; Computer science","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":true,"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":[0.04424261,0.001480653,0.003064933,0.004054717,0.001491688,0.004195589,0.003004031,0.001459113,0.005951955],"category_scores_gemma":[0.1225964,0.001130787,0.003236222,0.003779801,0.001769483,0.003650807,0.004613908,0.002879618,0.0006198768],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002483283,"about_ca_system_score_gemma":0.008802339,"about_ca_topic_candidate":true,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007614332,"about_ca_topic_score_gemma":0.008476177,"domain_scores_codex":[0.9757547,0.01686303,0.001357298,0.002632592,0.002662985,0.0007293077],"domain_scores_gemma":[0.9180627,0.06496577,0.001839497,0.01016012,0.004280607,0.0006912609],"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":[0.005815241,0.0006095465,0.01439029,0.001542606,0.002188011,0.0006960283,0.002055535,0.2091995,0.005216904,0.245476,0.01080028,0.5020101],"study_design_scores_gemma":[0.0004374531,0.0004478806,0.001974344,0.0002334114,0.0008872931,0.0001754777,0.0005143517,0.5339448,0.004516741,0.4497792,0.006987009,0.0001019925],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.0206791,0.0005589364,0.9723078,0.001340481,0.00006638781,0.001139119,0.00195866,0.0005614101,0.001388137],"genre_scores_gemma":[0.4373911,0.0004718196,0.5536957,0.0005809095,0.0001635619,0.001851397,0.003710427,0.0001986883,0.00193637],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9923857,"threshold_uncertainty_score":0.2339801,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1118521193370718,"score_gpt":0.3174271461467318,"score_spread":0.20557502680966,"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."}}