{"id":"W4413057136","doi":"10.2139/ssrn.5355709","title":"The Actual Performance of AI/ML Models in Predicting Radiation-Induced Toxicity in Head and Neck Cancer: A Systematic Review and Meta-Analysis","year":2025,"lang":"en","type":"preprint","venue":"SSRN Electronic Journal","topic":"Radiomics and Machine Learning in Medical Imaging","field":"Medicine","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"Princess Margaret Cancer Centre","funders":"","keywords":"Meta-analysis; Head and neck cancer; Toxicity; Oncology; Head and neck; Medicine; Cancer; Internal medicine; Surgery","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":[0.05248459,0.003920543,0.01892312,0.004753431,0.0007609421,0.00586674,0.003560144,0.004130478,0.00311324],"category_scores_gemma":[0.09699001,0.002907252,0.07355636,0.005496865,0.001493707,0.005130841,0.002439015,0.005428514,0.000633339],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00185538,"about_ca_system_score_gemma":0.001727713,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004060514,"about_ca_topic_score_gemma":0.007045588,"domain_scores_codex":[0.9612468,0.02238157,0.007922325,0.005462397,0.002449463,0.0005373983],"domain_scores_gemma":[0.9008995,0.08521046,0.006315141,0.004896372,0.002205994,0.0004724487],"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":[0.003253633,0.00001661241,0.007043844,0.04454192,0.9357606,0.00004771113,0.00005526069,0.00117405,0.0002049332,0.000160467,0.0002676327,0.00747333],"study_design_scores_gemma":[0.0006541035,0.0002042174,0.002689449,0.00216192,0.9923882,0.00004801695,0.00002738962,0.0006381534,0.0001059584,0.0004923831,0.000565063,0.00002511879],"study_design_candidate":"meta_analysis","study_design_consensus":"meta_analysis","genre_codex":"review","genre_gemma":"review","genre_scores_codex":[0.01637996,0.9759316,0.005037544,0.0004190383,0.0004718551,0.0001758244,0.0009855744,0.0001118573,0.0004868714],"genre_scores_gemma":[0.7618451,0.2230022,0.008897074,0.002063395,0.0009904496,0.0006511892,0.001631195,0.0002589293,0.0006604235],"genre_candidate":"review","genre_consensus":"review","teacher_disagreement_score":0.05248459,"threshold_uncertainty_score":0.2775683,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02091664877132908,"score_gpt":0.3266452240829061,"score_spread":0.305728575311577,"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."}}