{"id":"W4226034463","doi":"10.1007/s11548-022-02597-0","title":"Can uncertainty estimation predict segmentation performance in ultrasound bone imaging?","year":2022,"lang":"en","type":"article","venue":"International Journal of Computer Assisted Radiology and Surgery","topic":"Medical Imaging and Analysis","field":"Engineering","cited_by":7,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of British Columbia","funders":"Canadian Network for Research and Innovation in Machining Technology, Natural Sciences and Engineering Research Council of Canada; Orthopaedic Trauma Association","keywords":"Segmentation; Computer science; Ground truth; Artificial intelligence; Calibration; Artificial neural network; Image segmentation; Pattern recognition (psychology); Machine learning; Computer vision; Statistics; Mathematics","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":[0.0005892919,0.00007427099,0.0002026235,0.0003988448,0.00004778253,0.00002750089,0.0001039096,0.00001989734,0.00004395631],"category_scores_gemma":[0.00003611157,0.00007056935,0.00006568298,0.0001136522,0.00004468125,0.0001195153,0.0000198783,0.0002565995,3.498121e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001320836,"about_ca_system_score_gemma":0.0000394725,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000008036913,"about_ca_topic_score_gemma":0.000001168639,"domain_scores_codex":[0.99907,0.0001095838,0.0004130391,0.00007533724,0.0002299442,0.0001021308],"domain_scores_gemma":[0.9993604,0.0003680682,0.0001217389,0.0000406683,0.00006193058,0.00004716478],"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":[0.00003039213,0.00004442272,0.3514684,0.0000150893,0.0002043139,0.0002222997,0.0002963383,0.5023428,0.0003901801,0.00001521171,0.003709297,0.1412612],"study_design_scores_gemma":[0.0004205733,0.00001829866,0.1864292,0.00004646812,0.00002189086,0.003312686,0.00006069092,0.8086472,0.00003953578,0.0001117622,0.0007976272,0.00009413755],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.969627,0.0004823535,0.02739569,0.0009153883,0.001494335,0.00002118912,0.000006969193,0.00002068006,0.0000363761],"genre_scores_gemma":[0.9983154,0.0001860744,0.000985506,0.000270042,0.0001800456,0.000003650016,0.00004700987,0.000005790097,0.000006472773],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3063044,"threshold_uncertainty_score":0.2877735,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.006741805957928946,"score_gpt":0.2218219572577179,"score_spread":0.215080151299789,"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."}}