{"id":"W2772723798","doi":"10.1001/jama.2017.14585","title":"Diagnostic Assessment of Deep Learning Algorithms for Detection of Lymph Node Metastases in Women With Breast Cancer","year":2017,"lang":"en","type":"article","venue":"JAMA","topic":"AI in cancer detection","field":"Computer Science","cited_by":3297,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Medicine; Lymph node; Algorithm; H&E stain; Receiver operating characteristic; Deep learning; Test set; Artificial intelligence; Breast cancer; Lymph; Machine learning; Pathology; Cancer; Radiology; Staining; Internal medicine; Computer science","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":[0.01050773,0.0006758753,0.0003231566,0.001041312,0.0002644672,0.0008987148,0.0009089282,0.001271481,0.001463106],"category_scores_gemma":[0.05531484,0.0003145945,0.0005873727,0.0003030058,0.0004610012,0.0007069953,0.001115211,0.0006307684,0.0005283698],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000908552,"about_ca_system_score_gemma":0.0007039872,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00158678,"about_ca_topic_score_gemma":0.002571525,"domain_scores_codex":[0.9954594,0.002600079,0.0002995175,0.0005666784,0.0008800019,0.0001942985],"domain_scores_gemma":[0.981295,0.01230939,0.0019241,0.0008114295,0.00293362,0.0007264873],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.005770006,0.001467117,0.7942141,0.0003235688,0.0004184114,0.0002585937,0.0006587552,0.02430966,0.006019455,0.0003988116,0.003827885,0.1623336],"study_design_scores_gemma":[0.0006097928,0.008971066,0.6081914,0.0002892108,0.0003462187,0.001478432,0.001206664,0.3490835,0.02232084,0.002553859,0.004804628,0.000144329],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9899726,0.0006664116,0.006247954,0.0005475336,0.00004538289,0.0002103838,0.0003676377,0.0001100442,0.001831895],"genre_scores_gemma":[0.9913349,0.0001597133,0.007391129,0.0001079912,0.00001837687,0.00009202107,0.0005475278,0.00001427704,0.0003341345],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01050773,"threshold_uncertainty_score":0.05557084,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01440044758837548,"score_gpt":0.2921606175836773,"score_spread":0.2777601699953018,"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."}}