{"id":"W4415312233","doi":"10.2196/63908","title":"Clinical Information Extraction From Notes of Veterans With Lymphoid Malignancies: Natural Language Processing Study","year":2025,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Topic Modeling","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Center for Advancing Translational Sciences; National Cancer Institute","keywords":"Veterans Affairs; Information extraction; Pipeline (software); Information system; Natural language; Data extraction; Work (physics)","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00438788,0.0005947503,0.0004180908,0.003139481,0.0006573557,0.001295839,0.00111336,0.001039101,0.001245166],"category_scores_gemma":[0.02356069,0.0002233367,0.0009042412,0.001769557,0.0005745274,0.0009964197,0.001455734,0.001122536,0.0007419848],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001152173,"about_ca_system_score_gemma":0.002789053,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008452594,"about_ca_topic_score_gemma":0.01215332,"domain_scores_codex":[0.9968711,0.0009829066,0.0006272995,0.000794411,0.0005744786,0.0001498201],"domain_scores_gemma":[0.9708667,0.02243414,0.002131928,0.001345104,0.002782762,0.000439313],"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":[0.00155624,0.00172366,0.4270869,0.003811724,0.0003952021,0.009589113,0.007742765,0.01146695,0.03054076,0.001783725,0.03386455,0.4704383],"study_design_scores_gemma":[0.0007799704,0.001589665,0.4914587,0.001933441,0.0009555814,0.0160636,0.02301743,0.2724624,0.08805907,0.009882494,0.09331568,0.0004820262],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8885065,0.002277523,0.0647038,0.002634005,0.0002076418,0.002762223,0.03368853,0.002577106,0.002642786],"genre_scores_gemma":[0.7554902,0.001023296,0.1804724,0.001197166,0.0001704238,0.001520042,0.05873793,0.0001386149,0.001249904],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.008452594,"threshold_uncertainty_score":0.02320564,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01644192540689026,"score_gpt":0.3441439778989147,"score_spread":0.3277020524920244,"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."}}