{"id":"W4280613227","doi":"10.2196/34436","title":"The Value of Extracting Clinician-Recorded Affect for Advancing Clinical Research on Depression: Proof-of-Concept Study Applying Natural Language Processing to Electronic Health Records","year":2022,"lang":"en","type":"article","venue":"JMIR Formative Research","topic":"Mental Health via Writing","field":"Psychology","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Health Services Research and Development; U.S. Department of Veterans Affairs","keywords":"Mood; Affect (linguistics); Depression (economics); Vocabulary; Natural language processing; Reliability (semiconductor); Artificial intelligence; Set (abstract data type); Mood disorders; Medicine; Computer science; Psychology; Clinical psychology; Information retrieval; Psychiatry; Communication; Linguistics; Anxiety","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0530208,0.0009646419,0.0006614982,0.0009350146,0.0007070336,0.002514384,0.001816396,0.001411163,0.00181115],"category_scores_gemma":[0.1185937,0.000371369,0.001133014,0.0005605411,0.001777626,0.0033712,0.002175058,0.00112745,0.0005406147],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009143217,"about_ca_system_score_gemma":0.002826856,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001228295,"about_ca_topic_score_gemma":0.0009134837,"domain_scores_codex":[0.9725437,0.01867164,0.001964996,0.002364189,0.004002864,0.0004525703],"domain_scores_gemma":[0.7959157,0.1666415,0.006710826,0.008783975,0.02062352,0.001324409],"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":[0.007841363,0.009842389,0.1113719,0.01096162,0.001277359,0.002211843,0.0117836,0.008442109,0.2114198,0.004528201,0.007968504,0.6123513],"study_design_scores_gemma":[0.004467835,0.04596301,0.1826421,0.003177416,0.003936628,0.007693209,0.008465162,0.12135,0.5624132,0.01088428,0.04839601,0.0006112343],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6911898,0.002268961,0.2863263,0.002417801,0.000472688,0.01089808,0.001945058,0.001283717,0.003197514],"genre_scores_gemma":[0.5424426,0.001013325,0.4492119,0.000777801,0.0002045662,0.004076445,0.001539016,0.0001257672,0.0006085914],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.0530208,"threshold_uncertainty_score":0.2804041,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1960823211756976,"score_gpt":0.6268721470905724,"score_spread":0.4307898259148748,"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."}}