{"id":"W4395044957","doi":"10.2196/50428","title":"Examining Linguistic Differences in Electronic Health Records for Diverse Patients With Diabetes: Natural Language Processing Analysis","year":2024,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Health Literacy and Information Accessibility","field":"Health Professions","cited_by":13,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Rice University","keywords":"Health records; Natural language processing; Computer science; Linguistics; Linguistic analysis; Natural language; Diabetes mellitus; Artificial intelligence; Medicine; Health care","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":[0.002737547,0.0001977853,0.000523225,0.0005012882,0.0004209973,0.00007331423,0.0002607151,0.0001938273,0.0002561105],"category_scores_gemma":[0.00102134,0.0001294255,0.00006301965,0.001306617,0.00007218095,0.0009190606,0.00008025485,0.001273785,0.00003251748],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004719738,"about_ca_system_score_gemma":0.001800261,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00005556403,"about_ca_topic_score_gemma":0.0003601027,"domain_scores_codex":[0.9958167,0.0001975341,0.002213565,0.0001701319,0.0006760602,0.0009259902],"domain_scores_gemma":[0.9978315,0.000854135,0.0005621865,0.0002120623,0.0002234746,0.0003166035],"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":[0.00005915595,0.0000759287,0.7040853,0.00780065,0.00007558252,7.102184e-7,0.1114146,0.00000229819,2.181271e-8,0.0001508975,0.0006431229,0.1756917],"study_design_scores_gemma":[0.001833389,0.0004463578,0.30222,0.003802062,0.00008107755,1.840399e-7,0.0254075,0.6539609,2.985134e-7,0.00007683113,0.01181606,0.0003553288],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9903967,0.0004908163,0.006049174,0.0002130137,0.0004359706,0.001468464,0.00006348956,0.0001871562,0.0006951711],"genre_scores_gemma":[0.9930874,0.000023821,0.001665069,0.004062707,0.00015696,0.0003958818,0.0004721918,0.00001312351,0.0001228933],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.6539586,"threshold_uncertainty_score":0.5534034,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03443305287128406,"score_gpt":0.4213797797736721,"score_spread":0.3869467269023881,"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."}}