{"id":"W2921108820","doi":"10.1117/12.2512819","title":"Reliability of computer-aided diagnosis tools with multi-center MR datasets: impact of training protocol","year":2019,"lang":"en","type":"article","venue":"Medical Imaging 2019: Computer-Aided Diagnosis","topic":"AI in cancer detection","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Foothills Medical Centre; Ontario Brain Institute; Calgary Laboratory Services; University of Calgary","funders":"","keywords":"Computer science; Protocol (science); Cross-validation; Artificial intelligence; Classifier (UML); Data set; CAD; Computer-aided diagnosis; Reliability (semiconductor); Data mining; Data collection; Set (abstract data type); Pattern recognition (psychology); Machine learning; Medicine; Statistics; Mathematics; Pathology","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.001862388,0.0007646655,0.001464379,0.0004749494,0.0001411061,0.0002810075,0.00299364,0.0002536445,0.0003191091],"category_scores_gemma":[0.0004949183,0.000608828,0.0005286944,0.001196016,0.0006030347,0.002221081,0.001527813,0.0008120701,0.00007397561],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004653622,"about_ca_system_score_gemma":0.0007618347,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001015105,"about_ca_topic_score_gemma":0.00003092986,"domain_scores_codex":[0.9928449,0.0006321872,0.001730648,0.001733955,0.00198758,0.001070756],"domain_scores_gemma":[0.9925418,0.002546357,0.001106432,0.002587368,0.000537198,0.0006808176],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001929545,0.003712177,0.4980934,0.000738577,0.0003264105,0.00009518077,0.001400335,0.00362548,0.000150107,0.00009228662,0.03034851,0.4612246],"study_design_scores_gemma":[0.01478452,0.003120409,0.1882853,0.004087225,0.00008264255,0.0001901452,0.00004930957,0.7746182,0.00515439,0.0001981671,0.007923574,0.001506073],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1754101,0.00005866086,0.797222,0.00134881,0.001409128,0.02359344,0.000405579,0.0004952141,0.00005713181],"genre_scores_gemma":[0.7406021,0.00005942901,0.235858,0.001085463,0.0005653795,0.02152459,0.0001703895,0.0001244019,0.00001023802],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.7709928,"threshold_uncertainty_score":0.9996363,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0261076682686942,"score_gpt":0.3195067222023485,"score_spread":0.2933990539336543,"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."}}