{"id":"W4402917235","doi":"10.1109/cvprw63382.2024.00408","title":"BMAD: Benchmarks for Medical Anomaly Detection","year":2024,"lang":"en","type":"article","venue":"","topic":"Anomaly Detection Techniques and Applications","field":"Computer Science","cited_by":69,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Anomaly detection; Computer science; Artificial intelligence","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":[],"consensus_categories":[],"category_scores_codex":[0.0002274659,0.00006412516,0.00005696674,0.0000864082,0.0001050306,0.000157684,0.0003270913,0.00008343114,0.0002505992],"category_scores_gemma":[0.0000205033,0.00005308817,0.00007660066,0.0003267658,0.00001906693,0.0002189173,0.00006931344,0.00008559418,0.00004660491],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002647948,"about_ca_system_score_gemma":0.00005508493,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002563892,"about_ca_topic_score_gemma":0.00003019112,"domain_scores_codex":[0.9992976,0.000008277058,0.0001395361,0.0002752989,0.0001511038,0.0001282493],"domain_scores_gemma":[0.9995623,0.00008528926,0.00001307434,0.0002280462,0.00003369125,0.00007763823],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000001001047,0.00001551579,0.000007452336,0.00001547305,0.000007796965,0.000001977,0.00002576031,0.000001123924,0.001081705,0.4247597,0.008326851,0.5657556],"study_design_scores_gemma":[0.00006516619,0.0001183821,0.0001266914,0.00001390983,0.000004787575,0.00005601115,0.00000820995,0.4682869,0.04264506,0.02869313,0.4598421,0.000139703],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0007846026,0.0000804357,0.9839437,0.002533109,0.0002512554,0.0001915849,0.000001174513,0.000947548,0.01126657],"genre_scores_gemma":[0.9402639,0.0000170384,0.0572996,0.000336557,0.0001502422,0.0002770369,0.000002605138,0.000006435186,0.001646595],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9394793,"threshold_uncertainty_score":0.2743886,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.006793689405046456,"score_gpt":0.2734587328745697,"score_spread":0.2666650434695232,"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."}}