{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.009018477,0.00172926,0.0006178722,0.001162809,0.0006750312,0.001718171,0.001737021,0.001685537,0.02696878],"category_scores_gemma":[0.04652575,0.0009969805,0.0009727877,0.0005458195,0.0007218049,0.002007655,0.003818042,0.00215455,0.01598996],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001135814,"about_ca_system_score_gemma":0.001904373,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002258606,"about_ca_topic_score_gemma":0.002724502,"domain_scores_codex":[0.9946951,0.003383327,0.0003369107,0.0008090812,0.0006422016,0.0001334163],"domain_scores_gemma":[0.9782103,0.01607865,0.0008643767,0.002733974,0.001583472,0.00052928],"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":[0.003100849,0.0009309008,0.02075484,0.002348033,0.0003517811,0.001491491,0.009874248,0.05992491,0.03315246,0.01436867,0.2469861,0.6067157],"study_design_scores_gemma":[0.0005277973,0.000668766,0.005977354,0.0004927307,0.00009277422,0.001016623,0.001886584,0.7387879,0.05442457,0.02422032,0.1716077,0.0002968915],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.04258397,0.0002811456,0.6511437,0.00161465,0.0004555393,0.001738359,0.01972355,0.2768748,0.005584206],"genre_scores_gemma":[0.2650831,0.0001609836,0.6776327,0.0008066922,0.0001339873,0.003494782,0.03575478,0.01165012,0.005282819],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.02696878,"threshold_uncertainty_score":0.09021956,"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."}}