{"id":"W4411300793","doi":"10.2196/75608","title":"Automated Data Harmonization in Clinical Research: Natural Language Processing Approach","year":2025,"lang":"en","type":"article","venue":"JMIR Formative Research","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Center for Research Resources; National Center for Advancing Translational Sciences; National Institute of Neurological Disorders and Stroke; National Institute of Diabetes and Digestive and Kidney Diseases; National Heart, Lung, and Blood Institute","keywords":"Preprint; Harmonization; Computer science; Natural (archaeology); Natural language processing; Artificial intelligence; World Wide Web; History; Art; Archaeology","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":["research_integrity"],"consensus_categories":[],"category_scores_codex":[0.01823213,0.0001348847,0.0002523727,0.001374782,0.0005872656,0.0005351322,0.003816997,0.0001705662,0.000007430677],"category_scores_gemma":[0.002818623,0.000115588,0.00003020261,0.006068967,0.0003218696,0.00182209,0.00393008,0.003581174,0.0001022721],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002862509,"about_ca_system_score_gemma":0.0009545264,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002611982,"about_ca_topic_score_gemma":0.00004945644,"domain_scores_codex":[0.9915626,0.004578522,0.0006789174,0.0007910002,0.001447422,0.0009415479],"domain_scores_gemma":[0.9959728,0.001331665,0.00009405907,0.001559726,0.0009148075,0.000126971],"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":[0.00009565597,0.0006778823,0.02117053,0.001549198,0.00002447495,0.00005557118,0.0366128,0.0001639173,0.00005444605,0.01475274,0.0365078,0.888335],"study_design_scores_gemma":[0.0004276403,0.0000733423,0.07185178,0.0002286613,4.260964e-7,0.000003855385,0.001605338,0.9236751,0.00002626044,0.0003549964,0.001656371,0.00009629632],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5820723,0.0117088,0.24103,0.03732817,0.001947455,0.009857408,0.00007401328,0.004432619,0.1115492],"genre_scores_gemma":[0.9865137,0.00002723459,0.01232629,0.0001046746,0.00006794037,0.0001258026,0.0001168339,0.00001227254,0.0007051948],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9235111,"threshold_uncertainty_score":0.9987176,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2684285756844088,"score_gpt":0.5829767339268984,"score_spread":0.3145481582424896,"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."}}