{"id":"W4392283037","doi":"10.1038/s41467-024-46142-w","title":"Empirical data drift detection experiments on real-world medical imaging data","year":2024,"lang":"en","type":"article","venue":"Nature Communications","topic":"COVID-19 diagnosis using AI","field":"Medicine","cited_by":88,"is_retracted":false,"has_abstract":true,"ca_institutions":"University Health Network; York University; Trillium Health Centre; University of Toronto; Vector Institute","funders":"","keywords":"Computer science; Concept drift; Proxy (statistics); Data mining; Tracking (education); Artificial intelligence; Data stream mining; Machine learning","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":[0.02421142,0.001167575,0.0006807398,0.001745101,0.0006416741,0.001086948,0.001532751,0.001805184,0.0006462449],"category_scores_gemma":[0.07141264,0.0003056186,0.0007289839,0.001393153,0.001634974,0.001755721,0.001480885,0.001830761,0.0003696407],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001133652,"about_ca_system_score_gemma":0.0009213691,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003206742,"about_ca_topic_score_gemma":0.00310073,"domain_scores_codex":[0.9901541,0.004909599,0.0009793245,0.001876422,0.001684958,0.0003956116],"domain_scores_gemma":[0.9256805,0.05582481,0.004308908,0.006963972,0.00601801,0.001203795],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.00419671,0.003751952,0.2033152,0.001912743,0.001303873,0.0008976642,0.002194464,0.4575462,0.03463238,0.005767616,0.02806061,0.2564206],"study_design_scores_gemma":[0.0003022053,0.002170617,0.05169904,0.000216099,0.0001407389,0.0008952579,0.0006090723,0.8954087,0.03703604,0.005212246,0.006198462,0.0001115604],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9133402,0.002589169,0.0759365,0.001505717,0.000388387,0.0007169941,0.001847885,0.001786119,0.001888996],"genre_scores_gemma":[0.9158467,0.0004172424,0.07837599,0.00049214,0.00009168611,0.0002457012,0.003622242,0.0001370566,0.0007712208],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.02421142,"threshold_uncertainty_score":0.1280438,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1590621346926188,"score_gpt":0.5059469993638447,"score_spread":0.3468848646712259,"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."}}