{"id":"W4414137493","doi":"10.3389/fdgth.2025.1625444","title":"Synthetic patient and interview transcript creator: an essential tool for LLMs in mental health","year":2025,"lang":"en","type":"article","venue":"Frontiers in Digital Health","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"Natural Sciences and Engineering Research Council of Canada; Canadian Institutes of Health Research; Deutsches Krebsforschungszentrum; Nvidia","keywords":"Interview; Mental health; Software deployment; Diversity (politics); Population; Data collection; Patient data","routes":{"ca_aff":true,"ca_fund":true,"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":[0.0007846282,0.000220759,0.0005170626,0.0004137118,0.0001662201,0.0002156344,0.0004981809,0.00007318227,0.000001579711],"category_scores_gemma":[0.0000679357,0.0002279629,0.00006585704,0.0004474486,0.00006622839,0.0007260472,0.0001368489,0.0002941813,8.378432e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006127378,"about_ca_system_score_gemma":0.0006162708,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00054641,"about_ca_topic_score_gemma":0.0003787494,"domain_scores_codex":[0.9973623,0.0002601514,0.0007773967,0.0007101421,0.0002139551,0.0006760493],"domain_scores_gemma":[0.9990617,0.00006452989,0.0001526361,0.0004639078,0.00003106038,0.0002261743],"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":[0.00006030723,0.0002741004,0.03328444,0.001263975,0.000008713931,0.000003218529,0.006984097,0.00004123295,5.100537e-7,0.00613902,0.005239592,0.9467008],"study_design_scores_gemma":[0.01197905,0.01229758,0.09229981,0.01033296,0.00001317231,0.0001027242,0.009544999,0.5888628,0.00009626573,0.06243077,0.2093887,0.002651208],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1938392,0.01679174,0.7332206,0.04541023,0.00508574,0.004538645,0.0002460226,0.0002948557,0.0005729083],"genre_scores_gemma":[0.9663799,0.0003542601,0.02927594,0.0036826,0.00002469033,0.0001212176,0.00005401354,0.0000207598,0.00008656235],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9440496,"threshold_uncertainty_score":0.9296055,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01053373439376883,"score_gpt":0.2944215075587798,"score_spread":0.283887773165011,"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."}}