{"id":"W4417457716","doi":"10.1177/21501319251404193","title":"Balancing Model Complexity and Clinical Deployability in Deep Learning for Sociodemographic Information Extraction","year":2025,"lang":"en","type":"article","venue":"Journal of Primary Care & Community Health","topic":"Topic Modeling","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"North York General Hospital; University Health Network; University of Toronto","funders":"Canadian Institutes of Health Research","keywords":"Documentation; Deep learning; Convolutional neural network; Information overload; Artificial neural network; Binary classification; Health care; Unstructured data; Medical record; Data extraction","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.008947396,0.001625897,0.0009095416,0.001024511,0.0006550249,0.002141031,0.001468943,0.001216957,0.001276453],"category_scores_gemma":[0.02112826,0.000692991,0.0009869768,0.0009289569,0.0007628414,0.003346711,0.001956425,0.002303761,0.0004711415],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002026311,"about_ca_system_score_gemma":0.002372844,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01652278,"about_ca_topic_score_gemma":0.019914,"domain_scores_codex":[0.9978215,0.001023825,0.0002302468,0.0004872922,0.0002330722,0.0002040875],"domain_scores_gemma":[0.9913315,0.006546385,0.0004096502,0.0007698706,0.0007901276,0.0001525611],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001425918,0.0004494576,0.06287368,0.0003657464,0.0006769,0.0002459853,0.0005096098,0.6305141,0.006387588,0.003146808,0.00363133,0.2897728],"study_design_scores_gemma":[0.00004255696,0.0001419148,0.002734647,0.00004065161,0.00008508052,0.00003560256,0.00008144535,0.9898274,0.00245859,0.004023497,0.0005126547,0.00001597943],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7869869,0.001769711,0.2016881,0.002677347,0.0001186494,0.0003378931,0.001139925,0.002394802,0.002886641],"genre_scores_gemma":[0.9441062,0.000300922,0.0527249,0.0003012129,0.00003972248,0.0002396221,0.00142508,0.00006542145,0.0007970378],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01652278,"threshold_uncertainty_score":0.04731894,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06077800822176373,"score_gpt":0.3738910076647382,"score_spread":0.3131129994429745,"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."}}