{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01765072,0.001268956,0.001104883,0.001848318,0.002074284,0.004911074,0.003418169,0.002964315,0.004302815],"category_scores_gemma":[0.04430061,0.001225999,0.001765179,0.0007318485,0.00415448,0.006686783,0.006857722,0.004146961,0.001848693],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002703803,"about_ca_system_score_gemma":0.004398707,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006728405,"about_ca_topic_score_gemma":0.005749028,"domain_scores_codex":[0.9875295,0.007464848,0.0007426862,0.001529469,0.002019447,0.0007139844],"domain_scores_gemma":[0.9691697,0.02131086,0.00159702,0.003957228,0.002913118,0.001052023],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"qualitative","study_design_scores_codex":[0.0005239417,0.0009836789,0.006805682,0.001536431,0.0002158875,0.001314898,0.01442545,0.1301521,0.02634103,0.5807635,0.01109063,0.2258467],"study_design_scores_gemma":[0.00006999683,0.0002324998,0.0008811583,0.0002526377,0.00007256728,0.0004678154,0.001386591,0.8195269,0.008933836,0.141904,0.02615356,0.0001184721],"study_design_candidate":"qualitative","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.00851809,0.0002010778,0.9843776,0.0008344632,0.00005441944,0.0003475645,0.00007876995,0.003210649,0.002377317],"genre_scores_gemma":[0.2893741,0.0001951843,0.7046614,0.0004632591,0.00007654673,0.000779018,0.0004296271,0.0009781583,0.003042832],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01765072,"threshold_uncertainty_score":0.09334707,"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."}}