{"database":"biostudies-literature","file_versions":[],"scores":null,"additional":{"submitter":["Jha N"],"funding":["Korea Health Technology R&D Project through the Korea Health Industry Development Institute, funded by the Ministry of Health & Welfare, Republic of Korea","National Research Foundation of Korea"],"pagination":["20590"],"full_dataset_link":["https://www.ebi.ac.uk/biostudies/studies/S-EPMC9709043"],"repository":["biostudies-literature"],"omics_type":["Unknown"],"volume":["12(1)"],"pubmed_abstract":["The aim of this study was to develop an auto-segmentation algorithm for mandibular condyle using the 3D U-Net and perform a stress test to determine the optimal dataset size for achieving clinically acceptable accuracy. 234 cone-beam computed tomography images of mandibular condyles were acquired from 117 subjects from two institutions, which were manually segmented to generate the ground truth. Semantic segmentation was performed using basic 3D U-Net and a cascaded 3D U-Net. A stress test was performed using different sets of condylar images as the training, validation, and test datasets. Relative accuracy was evaluated using dice similarity coefficients (DSCs) and Hausdorff distance (HD). In the five stages, the DSC ranged 0.886-0.922 and 0.912-0.932 for basic 3D U-Net and cascaded 3D U-"],"journal":["Scientific reports"],"pubmed_title":["Fully automated condyle segmentation using 3D convolutional neural networks."],"pmcid":["PMC9709043"],"funding_grant_id":["2022R1H1A201117211","HI18C2383"],"pubmed_authors":["Kim T","Baek SH","Jha N","Ham S","Sung SJ","Kim YJ","Kim N"],"additional_accession":[]},"is_claimable":false,"name":"Fully automated condyle segmentation using 3D convolutional neural networks.","description":"The aim of this study was to develop an auto-segmentation algorithm for mandibular condyle using the 3D U-Net and perform a stress test to determine the optimal dataset size for achieving clinically acceptable accuracy. 234 cone-beam computed tomography images of mandibular condyles were acquired from 117 subjects from two institutions, which were manually segmented to generate the ground truth. Semantic segmentation was performed using basic 3D U-Net and a cascaded 3D U-Net. A stress test was performed using different sets of condylar images as the training, validation, and test datasets. Relative accuracy was evaluated using dice similarity coefficients (DSCs) and Hausdorff distance (HD). In the five stages, the DSC ranged 0.886-0.922 and 0.912-0.932 for basic 3D U-Net and cascaded 3D U-","dates":{"release":"2022-01-01T00:00:00Z","publication":"2022 Nov","modification":"2026-06-03T10:35:24.952Z","creation":"2025-04-05T09:43:50.33Z"},"accession":"S-EPMC9709043","cross_references":{"pubmed":["36446860"],"doi":["10.1038/s41598-022-24164-y"]}}