{"id":"W3046293795","doi":"10.2196/17784","title":"Identifying and Predicting Intentional Self-Harm in Electronic Health Record Clinical Notes: Deep Learning Approach","year":2020,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Suicide and Self-Harm Studies","field":"Psychology","cited_by":45,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Center for Advancing Translational Sciences; National Institute of Diabetes and Digestive and Kidney Diseases; National Institute on Drug Abuse; National Institute of Mental Health; National Institutes of Health; U.S. Department of Veterans Affairs","keywords":"Harm; Machine learning; Artificial intelligence; Word2vec; Leverage (statistics); Computer science; Deep learning; Convolutional neural network; Health care; Psychology; Medicine; Social psychology","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":[0.001799325,0.0001946721,0.000506323,0.0001114322,0.0001677723,0.00004775527,0.0002451616,0.0002174503,0.0002390851],"category_scores_gemma":[0.000733326,0.0001761784,0.0001017038,0.000340889,0.0001206655,0.0002108562,0.0002351311,0.001887138,0.0001119939],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008055606,"about_ca_system_score_gemma":0.0001544133,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004303544,"about_ca_topic_score_gemma":0.00002816455,"domain_scores_codex":[0.9967712,0.0002755111,0.001513675,0.000242432,0.0005420364,0.0006551782],"domain_scores_gemma":[0.9984485,0.0005754597,0.000369595,0.0001275731,0.00005039163,0.0004284996],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001674985,0.0004770969,0.5917087,0.0008544502,0.0004312709,0.00002525748,0.2153225,0.000010552,5.904169e-7,0.00180373,0.002635676,0.1865626],"study_design_scores_gemma":[0.01175521,0.002745023,0.3308459,0.0003066816,0.0001325249,0.0002612545,0.2621733,0.3585244,0.000001788667,0.000798386,0.03117223,0.001283302],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9755438,0.00110085,0.01503254,0.00253682,0.0004303245,0.0006046576,0.000002406092,0.0003450151,0.004403593],"genre_scores_gemma":[0.9869438,0.001236419,0.00393807,0.007236909,0.0004406163,0.0001065891,0.00004134978,0.00002527376,0.00003099234],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3585139,"threshold_uncertainty_score":0.819878,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07506992440623514,"score_gpt":0.4005812611141222,"score_spread":0.3255113367078871,"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."}}