{"id":"W4205096618","doi":"10.2196/25157","title":"Assessment of Natural Language Processing Methods for Ascertaining the Expanded Disability Status Scale Score From the Electronic Health Records of Patients With Multiple Sclerosis: Algorithm Development and Validation Study","year":2022,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Multiple Sclerosis Research Studies","field":"Medicine","cited_by":15,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto; Regent Park Community Health Centre; St. Michael's Hospital","funders":"St. Michael's Hospital Foundation; Li Ka Shing Foundation","keywords":"Expanded Disability Status Scale; F1 score; Recall; Artificial intelligence; Health records; Computer science; Convolutional neural network; Precision and recall; Standard score; Machine learning; Data mining; Medicine; Natural language processing; Algorithm; Multiple sclerosis; Health care; Psychology","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":true,"about_ca":true,"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":[0.01644672,0.001702817,0.00105044,0.002491664,0.0004739618,0.001582911,0.001902459,0.001759237,0.0007868846],"category_scores_gemma":[0.04079443,0.0004078716,0.001236634,0.001252682,0.0004494429,0.001670239,0.001148032,0.001270871,0.0003824366],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002127465,"about_ca_system_score_gemma":0.002526972,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01231946,"about_ca_topic_score_gemma":0.007366669,"domain_scores_codex":[0.9929739,0.003338888,0.0009744298,0.001319678,0.001156477,0.0002365648],"domain_scores_gemma":[0.9593505,0.03186536,0.001415702,0.001504556,0.005642904,0.0002208997],"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":[0.001856874,0.00268345,0.1228477,0.0008790905,0.001356086,0.0003873925,0.000385986,0.2252318,0.008296229,0.001106899,0.003949437,0.6310191],"study_design_scores_gemma":[0.0001640671,0.000517447,0.009682823,0.00006787806,0.0001221374,0.0002091713,0.0001052214,0.9840682,0.004098985,0.0004335374,0.0005100452,0.0000204605],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7128281,0.004406861,0.2721237,0.000874632,0.000176445,0.002078077,0.002297456,0.003499958,0.001714743],"genre_scores_gemma":[0.6400092,0.0009554427,0.3516532,0.0003180506,0.00005015102,0.00127563,0.005003711,0.0000917634,0.0006428912],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01644672,"threshold_uncertainty_score":0.08697963,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05208524982829123,"score_gpt":0.4109838042878863,"score_spread":0.3588985544595951,"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."}}