{"id":"W4417476627","doi":"10.1016/j.radonc.2025.111350","title":"The Actual Performance of ML/AI Models in Predicting Radiation-Induced Toxicity in Head and Neck Cancer: A Systematic Review and Meta-Analysis.","year":2025,"lang":"en","type":"review","venue":"Radiotherapy and Oncology","topic":"Radiomics and Machine Learning in Medical Imaging","field":"Medicine","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Princess Margaret Cancer Centre","funders":"Rijksuniversiteit Groningen","keywords":"Comparability; Reliability (semiconductor); Receiver operating characteristic; Predictive modelling; Head and neck; Systematic review; Meta-analysis; Head and neck cancer","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":[0.03955798,0.003812552,0.0173056,0.008869045,0.0006117174,0.005246476,0.003093094,0.002835226,0.002885847],"category_scores_gemma":[0.09336217,0.001806368,0.05801056,0.007548349,0.0009235198,0.003119894,0.001991508,0.00267933,0.0004713876],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00300732,"about_ca_system_score_gemma":0.005419392,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008235786,"about_ca_topic_score_gemma":0.01603585,"domain_scores_codex":[0.9796623,0.01058176,0.005559723,0.001731826,0.002124045,0.0003402776],"domain_scores_gemma":[0.9305218,0.05887086,0.006122488,0.001689818,0.002528481,0.0002665436],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"meta_analysis","study_design_gemma":"meta_analysis","study_design_scores_codex":[0.001083652,0.00002504323,0.005269686,0.4212004,0.5456103,0.00006569125,0.00007936855,0.001410036,0.0001356141,0.0002486458,0.0007628939,0.02410865],"study_design_scores_gemma":[0.0003209541,0.0001612809,0.00281794,0.04213543,0.9507713,0.00008056618,0.00003711461,0.0006143428,0.0001165072,0.0005426545,0.00237011,0.00003173515],"study_design_candidate":"meta_analysis","study_design_consensus":"meta_analysis","genre_codex":"review","genre_gemma":"review","genre_scores_codex":[0.001581994,0.9958501,0.001181921,0.0002828451,0.0001153271,0.0002444652,0.0005430924,0.00002838979,0.000171877],"genre_scores_gemma":[0.1146137,0.8746291,0.005636948,0.001254211,0.0003875613,0.001514575,0.001611302,0.00007329148,0.0002793639],"genre_candidate":"review","genre_consensus":"review","teacher_disagreement_score":0.03955798,"threshold_uncertainty_score":0.2092051,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04526372028538901,"score_gpt":0.4009531191821383,"score_spread":0.3556893988967493,"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."}}