{"id":"W4407686803","doi":"10.2196/66633","title":"Developing Effective Frameworks for Large Language Model–Based Medical Chatbots: Insights From Radiotherapy Education With ChatGPT","year":2025,"lang":"en","type":"article","venue":"JMIR Cancer","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":25,"is_retracted":false,"has_abstract":true,"ca_institutions":"Princess Margaret Cancer Centre; University of Toronto; University Health Network","funders":"","keywords":"Chatbot; Computer science; Reliability (semiconductor); Knowledge management; World Wide Web","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0001047219,0.0001342572,0.0002334097,0.000129217,0.0001232906,0.00001861483,0.00008044046,0.0003336831,0.0001404033],"category_scores_gemma":[0.000118498,0.0001042577,0.00004835234,0.0002764057,0.00003413295,0.00007097382,0.000007471914,0.0003271839,0.000003661905],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000554609,"about_ca_system_score_gemma":0.004193953,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0009208547,"about_ca_topic_score_gemma":0.001278136,"domain_scores_codex":[0.9990197,0.00003412125,0.000220076,0.0002924212,0.000217057,0.0002166146],"domain_scores_gemma":[0.9991015,0.0002600578,0.0000688971,0.0002177143,0.0002395401,0.0001123087],"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":[0.001906033,0.0007391194,0.01511352,0.0007846233,0.0002596902,0.000004831401,0.02301571,0.0002951589,0.0009434536,0.01579954,0.007343311,0.933795],"study_design_scores_gemma":[0.003762114,0.001313248,0.03452864,0.02048773,0.0004617317,0.000005974148,0.01982658,0.64455,0.1300666,0.03155222,0.1119045,0.001540618],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.4993218,0.005205133,0.4714713,0.02079799,0.0008233089,0.002089008,0.00001442453,0.00007306838,0.0002038695],"genre_scores_gemma":[0.962323,0.0001874639,0.009263228,0.0227313,0.0009430685,0.004041685,0.0001222188,0.00002622536,0.0003618155],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9322544,"threshold_uncertainty_score":0.7439892,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03320914112407512,"score_gpt":0.4544512183293075,"score_spread":0.4212420772052324,"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."}}