{"id":"W4377989430","doi":"10.1148/ryai.220232","title":"A Guide to Cross-Validation for Artificial Intelligence in Medical Imaging","year":2023,"lang":"en","type":"review","venue":"Radiology Artificial Intelligence","topic":"Radiomics and Machine Learning in Medical Imaging","field":"Medicine","cited_by":207,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"Society of Nuclear Medicine and Molecular Imaging","keywords":"Hyperparameter; Convolutional neural network; Artificial intelligence; Machine learning; Field (mathematics); Artificial neural network; Selection (genetic algorithm); Medicine; Computer science","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":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.0183092,0.002237676,0.002270718,0.005389956,0.000704458,0.002932875,0.003662393,0.004850091,0.01581334],"category_scores_gemma":[0.04155414,0.001532299,0.002053042,0.003950693,0.002940815,0.002953774,0.001652219,0.009095964,0.01575814],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001714998,"about_ca_system_score_gemma":0.004141749,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002802005,"about_ca_topic_score_gemma":0.003979466,"domain_scores_codex":[0.989458,0.005651167,0.001669551,0.0006810779,0.002410058,0.0001303108],"domain_scores_gemma":[0.9639122,0.02779038,0.001174559,0.001691452,0.005071135,0.0003603358],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00005605706,0.0001054865,0.0006034788,0.005223311,0.0002828856,0.0003785153,0.0003085568,0.009766768,0.001485135,0.09909843,0.3543152,0.5283762],"study_design_scores_gemma":[0.00005090665,0.000140833,0.001021257,0.004383193,0.00007115238,0.0009677153,0.00007501971,0.01303358,0.001132526,0.120718,0.8582732,0.0001325724],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.0004542219,0.1942613,0.7626147,0.01415432,0.004980525,0.0006196301,0.001634291,0.00310249,0.01817849],"genre_scores_gemma":[0.00636044,0.1162691,0.8405828,0.007707315,0.004509188,0.002656797,0.002318931,0.002114421,0.01748092],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9816908,"threshold_uncertainty_score":0.09682947,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1001407464111719,"score_gpt":0.4829447736101836,"score_spread":0.3828040271990117,"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."}}