{"id":"W2735666957","doi":"10.1007/s10278-017-9997-y","title":"Thyroid Nodule Classification in Ultrasound Images by Fine-Tuning Deep Convolutional Neural Network","year":2017,"lang":"en","type":"article","venue":"Journal of Digital Imaging","topic":"AI in cancer detection","field":"Computer Science","cited_by":389,"is_retracted":false,"has_abstract":true,"ca_institutions":"Royal University Hospital; University of Saskatchewan","funders":"Saskatchewan Health Research Foundation; Compute Canada; University of Saskatchewan","keywords":"Convolutional neural network; Nodule (geology); Computer science; Ultrasound; Radiology; Thyroid; Artificial intelligence; Medicine; Pattern recognition (psychology); Internal medicine; Geology","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":["scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.0003601072,0.0001204865,0.0001646539,0.0001021667,0.0002754033,0.001596186,0.0008872387,0.000027926,0.000003607581],"category_scores_gemma":[0.0002603777,0.0001168355,0.00008388932,0.0001261832,0.0001133822,0.006224265,0.0001351217,0.0003103774,0.000006746397],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001891982,"about_ca_system_score_gemma":0.00006780255,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0000086657,"about_ca_topic_score_gemma":0.000005203361,"domain_scores_codex":[0.9987494,0.00003049478,0.000413487,0.0001945295,0.0003414194,0.0002706976],"domain_scores_gemma":[0.9985273,0.000146735,0.000730311,0.0003333871,0.0001804496,0.000081766],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.00007945564,0.0001892734,0.6418296,0.00002491579,0.00006511698,0.0001581265,0.0005722557,0.01107401,0.06359628,0.001458621,0.02826627,0.2526861],"study_design_scores_gemma":[0.002519765,0.0002367037,0.5420917,0.0002867164,0.00002634294,0.002191848,0.0002009377,0.4234895,0.00473599,0.0194601,0.003981198,0.0007791697],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1958249,0.002141018,0.787783,0.006272598,0.002453815,0.000129127,0.000008908976,0.00006697842,0.005319707],"genre_scores_gemma":[0.9945939,0.00002119247,0.004811641,0.00008313859,0.0004100797,0.00000157428,0.00000198708,0.00001038943,0.00006609805],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7987691,"threshold_uncertainty_score":0.9994403,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01618639259369834,"score_gpt":0.2612800436195284,"score_spread":0.2450936510258301,"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."}}