{"id":"W3179608547","doi":"10.48550/arxiv.2107.04540","title":"Objective task-based evaluation of artificial intelligence-based medical imaging methods: Framework, strategies and role of the physician","year":2021,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Radiomics and Machine Learning in Medical Imaging","field":"Medicine","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"","keywords":"Computer science; Task (project management); Modalities; Context (archaeology); Artificial intelligence; Medical imaging; Focus (optics); Engineering; Systems engineering","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":[0.001832146,0.0002251592,0.0005351438,0.0001900615,0.00007850661,0.00003855492,0.0003172317,0.0002468313,0.00008794489],"category_scores_gemma":[0.001725944,0.000195389,0.0002669323,0.0004592848,0.0006904797,0.00007838418,0.0002722801,0.001282336,5.05211e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001389009,"about_ca_system_score_gemma":0.002750377,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003413431,"about_ca_topic_score_gemma":0.00002386011,"domain_scores_codex":[0.9975669,0.0008815586,0.0003468709,0.0005511436,0.0004567692,0.0001967371],"domain_scores_gemma":[0.9976836,0.000600044,0.0004388979,0.0005632055,0.0005789039,0.0001353695],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000373434,0.0007817737,0.01622025,0.001037912,0.0005862922,0.00009618556,0.00196282,0.6073627,0.007457729,0.05605246,0.00001041255,0.308058],"study_design_scores_gemma":[0.000302333,0.00003226306,0.002917465,0.001355987,0.0007966466,0.000002091244,0.00538175,0.9165397,0.006634994,0.06588144,0.000007332471,0.0001479516],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4864366,0.0004401451,0.5114173,0.0004941334,0.0002183376,0.0003236598,0.000005058114,0.00002137906,0.0006434564],"genre_scores_gemma":[0.9914489,0.00004015173,0.008103972,0.000264357,0.0000932418,0.000001832844,0.0000222427,0.0000211318,0.000004140142],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.5050124,"threshold_uncertainty_score":0.7967732,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05626120817646408,"score_gpt":0.3078303269020632,"score_spread":0.2515691187255992,"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."}}