{"id":"W4223430324","doi":"10.1038/s41746-022-00592-y","title":"Machine learning for medical imaging: methodological failures and recommendations for the future","year":2022,"lang":"en","type":"review","venue":"npj Digital Medicine","topic":"Radiomics and Machine Learning in Medical Imaging","field":"Medicine","cited_by":593,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University; Mila - Quebec Artificial Intelligence Institute; Institut National de la Recherche Scientifique","funders":"Novo Nordisk Fonden; Agence Nationale de la Recherche","keywords":"Medical imaging; Psychology; Computer science; Data science; Engineering ethics; Artificial intelligence; Medical physics; Medicine; Engineering","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":"codex-gemma-dda1882f352a","candidate_categories":["metaresearch","metaepi_narrow","research_integrity","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.003334709,0.0005751516,0.00227669,0.0002623149,0.0006411668,0.00009913933,0.0004637994,0.0002732984,0.001149923],"category_scores_gemma":[0.01993551,0.0003019882,0.000584203,0.0003350115,0.0005615414,0.00009845938,0.0002809275,0.002404811,0.000003074607],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001115296,"about_ca_system_score_gemma":0.0003046417,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002118673,"about_ca_topic_score_gemma":0.000001982397,"domain_scores_codex":[0.9966528,0.000291466,0.001014982,0.0007546689,0.0007451072,0.0005409374],"domain_scores_gemma":[0.9844901,0.01407577,0.0004473854,0.0003995963,0.00009075358,0.0004963703],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00006559939,0.00004911851,0.00002069104,0.003961235,0.0002909222,0.00002892279,0.00008144583,4.583312e-7,9.189181e-8,0.001841357,0.03404057,0.9596196],"study_design_scores_gemma":[0.001767496,0.0004877794,0.000003681505,0.003334088,0.001834351,0.001380518,0.0004025262,0.004588564,2.768323e-8,0.0005076544,0.9854048,0.0002885217],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"review","genre_gemma":"review","genre_scores_codex":[2.188141e-7,0.8506236,0.02131817,0.1236454,0.00123593,0.002004801,0.0000898205,0.0001311352,0.000950943],"genre_scores_gemma":[0.00001542303,0.9840838,0.003109542,0.002742042,0.004282169,0.001007613,0.002757158,0.00013874,0.001863565],"genre_candidate":"review","genre_consensus":"review","teacher_disagreement_score":0.959331,"threshold_uncertainty_score":0.9999432,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1037579113986433,"score_gpt":0.4339280365929593,"score_spread":0.330170125194316,"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."}}