{"id":"W4413821196","doi":"10.1016/j.jtho.2025.07.070","title":"PP01.21: Evaluating the Accuracy & Reproducibility of ChatGPT Models in Answering Patient’s Lung Cancer Queries","year":2025,"lang":"en","type":"article","venue":"Journal of Thoracic Oncology","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"Princess Margaret Cancer Centre; Trillium Health Centre; University of Toronto","funders":"","keywords":"Medicine; Reproducibility; Lung cancer; Cancer; Oncology; Medical physics; Internal medicine; Statistics","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"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":[0.04601778,0.002459398,0.001368756,0.003429031,0.001449087,0.003392307,0.003640441,0.00505271,0.003546358],"category_scores_gemma":[0.1248531,0.0008340818,0.002803931,0.001692279,0.001795178,0.002948642,0.003641898,0.002592418,0.003866736],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001596804,"about_ca_system_score_gemma":0.002655824,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01552362,"about_ca_topic_score_gemma":0.01187601,"domain_scores_codex":[0.9575517,0.02290442,0.005276717,0.00690269,0.006254602,0.001109792],"domain_scores_gemma":[0.8403905,0.1158507,0.005909601,0.01712234,0.01735243,0.003374601],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.01634318,0.00554868,0.5725159,0.00278597,0.01342839,0.001192496,0.002438158,0.06403318,0.01773556,0.001593781,0.08089421,0.2214905],"study_design_scores_gemma":[0.002565861,0.00959015,0.3195935,0.0003235718,0.003036118,0.00314503,0.001604964,0.5966658,0.03907137,0.003420072,0.02036449,0.0006190158],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9330959,0.001926854,0.02712074,0.00141879,0.001592482,0.0009363477,0.0210919,0.007620675,0.005196126],"genre_scores_gemma":[0.932567,0.0002153866,0.02903425,0.0006139165,0.0003645974,0.0005375603,0.03274499,0.0007317417,0.003190568],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.04601778,"threshold_uncertainty_score":0.2433682,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2456288239374304,"score_gpt":0.594969846062767,"score_spread":0.3493410221253367,"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."}}