{"id":"W4409880148","doi":"10.1016/j.eswa.2025.127928","title":"Natural language processing methods for assessing social determinants of health in the electronic health records: A narrative review","year":2025,"lang":"en","type":"review","venue":"Expert Systems with Applications","topic":"Mental Health via Writing","field":"Psychology","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"North York General Hospital; Toronto Western Hospital; University Health Network; University of Toronto","funders":"Canadian Institutes of Health Research; University of Toronto","keywords":"Health records; Narrative; Computer science; Natural language processing; Natural (archaeology); Social determinants of health; Data science; Electronic health record; Artificial intelligence; Linguistics; Public health; History; Medicine; Health care; Nursing; Political science","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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.005735903,0.0004318593,0.002910665,0.0002397713,0.0007188699,0.00004964051,0.0006050961,0.0001835647,0.000005183218],"category_scores_gemma":[0.0000702302,0.0002878063,0.000223089,0.001322672,0.00007244973,0.0000830389,0.00003635103,0.0006736175,0.000002656257],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009655906,"about_ca_system_score_gemma":0.003415677,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0009407543,"about_ca_topic_score_gemma":0.0002437309,"domain_scores_codex":[0.9917104,0.004292815,0.002269202,0.000713152,0.0001984119,0.0008159767],"domain_scores_gemma":[0.9949808,0.001183813,0.003063405,0.0005948744,0.0001141817,0.00006291586],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00000365628,0.00007595749,0.000001087963,0.1490042,0.00002706362,2.625949e-7,0.009509011,1.035783e-8,5.035091e-8,0.0005123451,0.001515178,0.8393512],"study_design_scores_gemma":[0.0002691404,0.0001066207,0.000004530143,0.1621104,0.00006923416,0.0000964473,0.02200587,0.00002306733,6.103991e-8,0.00001224124,0.8150431,0.0002592957],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"review","genre_gemma":"review","genre_scores_codex":[3.988232e-7,0.9596878,0.02190696,0.0008263624,0.0001776343,0.01692873,0.00005216809,0.00004673785,0.0003732326],"genre_scores_gemma":[0.0001090556,0.9411356,0.006333081,0.001315639,0.0002824203,0.05021675,0.0002602345,0.00006651288,0.000280725],"genre_candidate":"review","genre_consensus":"review","teacher_disagreement_score":0.8390918,"threshold_uncertainty_score":0.9999574,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1011913577653562,"score_gpt":0.5857522439506605,"score_spread":0.4845608861853043,"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."}}