{"id":"W4415473192","doi":"10.2196/78332","title":"Enabling Just-in-Time Clinical Oncology Analysis With Large Language Models: Feasibility and Validation Study Using Unstructured Synthetic Data","year":2025,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Topic Modeling","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Proof of concept; Clinical Oncology; Unstructured data; Synthetic data; Software; Patient data; Data format","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":[0.003695136,0.00014095,0.0005098694,0.0003119505,0.00008484535,0.0001415216,0.001249517,0.000194719,0.00001891484],"category_scores_gemma":[0.000412569,0.0001071882,0.00003945989,0.0009909382,0.00008343899,0.0009599949,0.001265025,0.0004403566,0.000001881008],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001069516,"about_ca_system_score_gemma":0.0004324911,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00005330774,"about_ca_topic_score_gemma":0.0001335159,"domain_scores_codex":[0.9973779,0.0002732497,0.001051772,0.0003526755,0.0006644847,0.0002799268],"domain_scores_gemma":[0.9978296,0.0003675678,0.0002210782,0.00135098,0.00007185536,0.0001589527],"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":[0.0003284028,0.005449186,0.3496388,0.001069,0.002649315,0.0004235769,0.1245751,0.03749438,0.00001506176,0.010705,0.0002406391,0.4674115],"study_design_scores_gemma":[0.001215978,0.00006539566,0.001945618,0.00005620497,0.0001582307,0.000006783217,0.003408025,0.992733,0.000002951258,0.0002714903,0.0000280665,0.0001082662],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5571305,0.00002663948,0.4421743,0.00009634169,0.000053794,0.000302025,0.000003976525,0.00004304135,0.000169365],"genre_scores_gemma":[0.9438596,0.000009913439,0.05567805,0.0003833151,0.00002547328,0.000007321587,0.0000213342,0.000003584637,0.00001135759],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9552386,"threshold_uncertainty_score":0.4371008,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1076523247402565,"score_gpt":0.4219480722279051,"score_spread":0.3142957474876487,"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."}}