{"id":"W4410777321","doi":"10.2196/71374","title":"Development of a Cohesive Predictive Model for Substance Use Disorder Rehabilitation Using Passive Digital Biomarkers, Psychological Assessments, and Automated Facial Emotion Recognition: Protocol for a Prospective Cohort Study","year":2025,"lang":"en","type":"article","venue":"JMIR Research Protocols","topic":"Emotion and Mood Recognition","field":"Psychology","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Preprint; Rehabilitation; Clinical psychology; Psychology; Emotion recognition; Artificial intelligence; Applied psychology; Computer science; Physical medicine and rehabilitation; Medicine; Physical therapy; World Wide Web","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001335974,0.0002569824,0.0003820338,0.0004558377,0.00046311,0.0001737232,0.0001511565,0.0002563015,0.00003057336],"category_scores_gemma":[0.0005990224,0.0002321742,0.00009586712,0.0006163041,0.0003095884,0.0005593426,0.00007051217,0.0002635539,0.00000307967],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003458068,"about_ca_system_score_gemma":0.0002895958,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000006255178,"about_ca_topic_score_gemma":0.00003941483,"domain_scores_codex":[0.9967909,0.0005183081,0.0007806006,0.0009287109,0.0005016539,0.0004798484],"domain_scores_gemma":[0.9964164,0.0007033528,0.000321065,0.0002746182,0.002186416,0.00009818484],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.08047877,0.0481007,0.6408452,0.003876439,0.002267536,0.000008560457,0.03012521,0.00009569339,0.002251513,0.001421808,0.009279396,0.1812492],"study_design_scores_gemma":[0.03566593,0.0131545,0.8707018,0.003234806,0.0000474536,0.000005018763,0.01108004,0.05360231,0.0004359154,0.009595054,0.001845757,0.0006314213],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"protocol","genre_gemma":"protocol","genre_scores_codex":[0.1308232,2.248202e-7,0.06820779,0.00003777819,0.00001538469,0.8002253,0.0002960673,0.0001403622,0.0002539008],"genre_scores_gemma":[0.04386313,3.705549e-8,0.02068763,0.000009196161,0.00002457764,0.9350204,0.000131092,0.00002916047,0.0002347423],"genre_candidate":"protocol","genre_consensus":"protocol","teacher_disagreement_score":0.2298566,"threshold_uncertainty_score":0.9467788,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2969792008486836,"score_gpt":0.577243984343511,"score_spread":0.2802647834948274,"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."}}