Application of UNETR for automatic cochlear segmentation in temporal bone CTs.
Automatic segmentation
Cochlear implant surgery
Deep learning
Temporal bone CT
Journal
Auris, nasus, larynx
ISSN: 1879-1476
Titre abrégé: Auris Nasus Larynx
Pays: Netherlands
ID NLM: 7708170
Informations de publication
Date de publication:
Apr 2023
Apr 2023
Historique:
received:
24
04
2022
revised:
02
06
2022
accepted:
30
06
2022
pubmed:
16
8
2022
medline:
21
3
2023
entrez:
15
8
2022
Statut:
ppublish
Résumé
To investigate the feasibility of a deep learning method based on a UNETR model for fully automatic segmentation of the cochlea in temporal bone CT images. The normal temporal bone CTs of 77 patients were used in 3D U-Net and UNETR model automatic cochlear segmentation. Tests were performed on two types of CT datasets and cochlear deformity datasets. Through training the UNETR model, when batch_size=1, the Dice coefficient of the normal cochlear test set was 0.92, which was higher than that of the 3D U-Net model; on the GE 256 CT, SE-DS CT and Cochlear Deformity CT dataset tests, the Dice coefficients were 0.91, 0.93, 0 93, respectively. According to the anatomical characteristics of the temporal bone, the use of the UNETR model can achieve fully automatic segmentation of the cochlea and obtain an accuracy close to manual segmentation. This method is feasible and has high accuracy.
Identifiants
pubmed: 35970625
pii: S0385-8146(22)00184-5
doi: 10.1016/j.anl.2022.06.008
pii:
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
212-217Informations de copyright
Copyright © 2022 Elsevier Ltd. All rights reserved.
Déclaration de conflit d'intérêts
Declaration of Competing Interest The authors declare that they have no conflicts of interest.