{"id":"W4409701140","doi":"10.2196/60439","title":"Data Collection for Automatic Depression Identification in Spanish Speakers Using Deep Learning Algorithms: Protocol for a Case-Control Study","year":2025,"lang":"en","type":"article","venue":"JMIR Research Protocols","topic":"Mental Health via Writing","field":"Psychology","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Preprint; Computer science; Identification (biology); Protocol (science); Data collection; Depression (economics); Artificial intelligence; Control (management); Algorithm; Natural language processing; Machine learning; Psychology; World Wide Web; Medicine; Statistics; Mathematics","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[{"model":"gemma","categories":[],"domain":null,"study_design":"observational","genre":"protocol","about_ca_system":false,"about_ca_topic":false,"confidence":"low","status":"direct model label, unvalidated"},{"model":"gpt","categories":[],"domain":null,"study_design":"observational","genre":"protocol","about_ca_system":false,"about_ca_topic":false,"confidence":"high","status":"direct model label, unvalidated"}],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01341584,0.001345314,0.001693856,0.001203284,0.001461236,0.0008896433,0.001415537,0.002003365,0.02987029],"category_scores_gemma":[0.01729147,0.0008011857,0.001213742,0.0009727636,0.001231271,0.0007679584,0.001228199,0.001731803,0.00673922],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001523765,"about_ca_system_score_gemma":0.00475897,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002275681,"about_ca_topic_score_gemma":0.002952119,"domain_scores_codex":[0.9946417,0.002551635,0.0009605115,0.0007647035,0.0007484767,0.0003328635],"domain_scores_gemma":[0.9918091,0.00155349,0.0008271994,0.001704232,0.003640188,0.0004658449],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"observational","study_design_scores_codex":[0.128813,0.02849207,0.03680589,0.02382119,0.0019049,0.005230348,0.003608559,0.006336608,0.0226715,0.01376291,0.3673688,0.3611842],"study_design_scores_gemma":[0.08226369,0.04909246,0.1198984,0.01381253,0.001643984,0.002928322,0.003118602,0.008615924,0.01748402,0.01450958,0.6857966,0.0008359358],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"protocol","genre_gemma":"protocol","genre_scores_codex":[0.02620137,0.001042566,0.04036376,0.001268378,0.001158448,0.9010488,0.02409525,0.0004210139,0.004400485],"genre_scores_gemma":[0.009881547,0.0002708246,0.01300916,0.0004038124,0.00008114919,0.9722946,0.002859919,0.0000266538,0.001172285],"genre_candidate":"protocol","genre_consensus":"protocol","teacher_disagreement_score":0.02987029,"threshold_uncertainty_score":0.09992611,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.4652356711600004,"score_gpt":0.6603056894238801,"score_spread":0.1950700182638797,"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."}}