{"id":"W4390837850","doi":"10.1038/s41598-024-51723-2","title":"AI improves accuracy, agreement and efficiency of pathologists for Ki67 assessments in breast cancer","year":2024,"lang":"en","type":"article","venue":"Scientific Reports","topic":"AI in cancer detection","field":"Computer Science","cited_by":37,"is_retracted":false,"has_abstract":true,"ca_institutions":"Vector Institute; St. Michael's Hospital; Princess Margaret Cancer Centre; University Health Network; Toronto Metropolitan University; University of Toronto","funders":"Mitacs","keywords":"Medicine; Breast cancer; Kappa; Concordance; Turnaround time; Reproducibility; Consistency (knowledge bases); Diagnostic accuracy; Cohen's kappa; Cancer; Medical physics; Internal medicine; Artificial intelligence; Machine learning; Computer science; Statistics","routes":{"ca_aff":true,"ca_fund":true,"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":[0.001619968,0.0001179593,0.0001558787,0.0002669198,0.0001075019,0.0004606597,0.0002569553,0.00005086025,0.00001174406],"category_scores_gemma":[0.00004417971,0.0001007022,0.000054568,0.0007840443,0.0001711824,0.0006283174,0.0002181717,0.00009548206,0.000001400694],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001595428,"about_ca_system_score_gemma":0.0003747017,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001415747,"about_ca_topic_score_gemma":0.00006604638,"domain_scores_codex":[0.9978963,0.00002879075,0.0004642459,0.0009274821,0.0004106424,0.0002725802],"domain_scores_gemma":[0.9989324,0.00006915558,0.0001921276,0.0006066927,0.0001436182,0.00005605042],"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":[0.00002369405,0.0002575004,0.01917312,0.0008803832,0.00004088195,0.0004566545,0.002193644,0.0005182481,0.2412588,0.001996559,0.01317668,0.7200239],"study_design_scores_gemma":[0.001169085,0.0005702195,0.1460713,0.001771858,0.00008978661,0.002075306,0.0003013749,0.4786123,0.2070957,0.1118547,0.04885471,0.001533614],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7034987,0.001935297,0.2628691,0.002498618,0.02740551,0.001263404,0.0000254098,0.0001850867,0.0003188323],"genre_scores_gemma":[0.9960814,0.00002504655,0.003271896,0.00004730595,0.00004256316,0.0001430538,0.000002761669,0.000007102148,0.0003789274],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.7184902,"threshold_uncertainty_score":0.4442152,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01751447108590855,"score_gpt":0.3350775172729329,"score_spread":0.3175630461870244,"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."}}