{"id":"W4321612729","doi":"10.1001/jamanetworkopen.2023.0524","title":"A Competition, Benchmark, Code, and Data for Using Artificial Intelligence to Detect Lesions in Digital Breast Tomosynthesis","year":2023,"lang":"en","type":"article","venue":"JAMA Network Open","topic":"Digital Radiography and Breast Imaging","field":"Medicine","cited_by":28,"is_retracted":false,"has_abstract":true,"ca_institutions":"McMaster University","funders":"National Institute of Biomedical Imaging and Bioengineering; National Cancer Institute","keywords":"Data set; Artificial intelligence; Benchmark (surveying); Sensitivity (control systems); Set (abstract data type); Computer science; Test set; Code (set theory); Health care; Tomosynthesis; Machine learning; Medical physics; Medicine; Mammography; Breast cancer; Cancer; Cartography","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":[0.0006589637,0.0001372666,0.0002840117,0.0001815352,0.0001347507,0.0006100695,0.0004300663,0.00005120329,0.00002475226],"category_scores_gemma":[0.0002613031,0.0001295421,0.00004487448,0.001209893,0.00006578769,0.0008357342,0.0007432314,0.0001044534,0.00001763658],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002680367,"about_ca_system_score_gemma":0.00006815275,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00006522721,"about_ca_topic_score_gemma":0.0001706851,"domain_scores_codex":[0.9987113,0.00002096634,0.0003075735,0.0004429911,0.0001406306,0.0003764851],"domain_scores_gemma":[0.9988744,0.0003867589,0.00005036159,0.0004459571,0.00006366867,0.0001788464],"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":[0.0007752729,0.00008545197,0.009798572,0.00005468513,0.00005918763,0.00006157946,0.00007924342,0.0007191089,0.00004737216,0.005457054,0.002271337,0.9805911],"study_design_scores_gemma":[0.002161909,0.001038708,0.3415365,0.01106077,0.0004897154,0.002466172,0.004023049,0.4353319,0.0005766157,0.1624992,0.0366335,0.002181918],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9052783,0.0002096516,0.06746934,0.01243597,0.0003465654,0.003759322,0.003072325,0.0001870711,0.007241419],"genre_scores_gemma":[0.9909766,0.00001679468,0.007804139,0.00035839,0.0003223041,0.00004173342,0.0004063599,0.00002465186,0.00004906212],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9784092,"threshold_uncertainty_score":0.5882914,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08941549269208739,"score_gpt":0.3487572924536095,"score_spread":0.2593417997615221,"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."}}