Evaluation of ultrasonic fibrosis diagnostic system using convolutional network for ordinal regression.
Deep learning
Liver fibrosis
Ultrasound image
Journal
International journal of computer assisted radiology and surgery
ISSN: 1861-6429
Titre abrégé: Int J Comput Assist Radiol Surg
Pays: Germany
ID NLM: 101499225
Informations de publication
Date de publication:
Nov 2021
Nov 2021
Historique:
received:
19
01
2021
accepted:
03
09
2021
pubmed:
22
9
2021
medline:
17
11
2021
entrez:
21
9
2021
Statut:
ppublish
Résumé
Diagnosis of liver fibrosis is important for establishing treatment and assessing the risk of carcinogenesis. Ultrasound imaging is an excellent diagnostic method as a screening test in terms of non-invasiveness and simplicity. The purpose of this study was to automatically diagnose liver fibrosis using ultrasound images to reduce the burden on physicians. We proposed and implemented a system for extracting regions of liver parenchyma utilizing U-Net. Using regions of interest, the stage of fibrosis was classified as F0, F1, F2, F3, or F4 utilizing CORALNet, an ordinal regression model based on ResNet18. The effectiveness of the proposed system was verified. The system implemented using U-Net had a maximum mean Dice coefficient of 0.929. The results of classification of liver fibrosis utilizing CORALNet had a mean absolute error (MAE) of 1.22 and root mean square error (RMSE) of 1.60. The per-case results had a MAE of 1.55 and RMSE of 1.34. U-Net extracted regions of liver parenchyma from the images with high accuracy, and CORALNet showed effectiveness using ordinal information to classify fibrosis in the images. As a future task, we will study a model that is less dependent on teaching data.
Identifiants
pubmed: 34545465
doi: 10.1007/s11548-021-02491-1
pii: 10.1007/s11548-021-02491-1
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
1969-1975Informations de copyright
© 2021. CARS.
Références
Virmani J, Kumar V, Kalra N, Khandelwal N (2013) SVM-based characterisation of liver cirrhosis by singular value decomposition of GLCM matrix. Int J Artif Intell Soft Comput 3(3):276–296
doi: 10.1504/IJAISC.2013.053407
Meng D, Zhang L, Cao G, Cao W, Zhang G, Hu B (2017) Liver fibrosis classification based on transfer learning and FCNet for ultrasound images. Ieee Access 5:5804–5810
Milletari F, Navab N, Ahmadi SA (2016) V-net: fully convolutional neural networks for volumetric medical image segmentation. In 2016 fourth international conference on 3D vision (3DV). IEEE, 565–571
Ronneberger O, Fischer P, Brox T (2015) U-net: convolutional networks for biomedical image segmentation. In international conference on medical image computing and computer-assisted intervention. Springer, Cham. 234–241
Li L, Lin HT (2006) Ordinal regression by extended binary classification. Adv Neural Inf Process Syst 19:865–872
Niu Z, Zhou M, Wang L, Gao X, Hua G (2016) Ordinal regression with multiple output cnn for age estimation. In proceedings of the IEEE conference on computer vision and pattern recognition, 4920–4928
Cao W, Mirjalili V, Raschka S (2020) Rank consistent ordinal regression for neural networks with application to age estimation. Pattern Recogn Lett 140:325–331
doi: 10.1016/j.patrec.2020.11.008
Masuzaki R, Tateishi R, Yoshida H, Goto E, Sato T, Ohki T, Goto T, Yoshida H, Kanai F, Sugioka Y, Ikeda H, Shiina S, Kawabe T, Omata M (2008) Comparison of liver biopsy and transient elastography based on clinical relevance. Can JGastroenterol 22(9):753–757. https://doi.org/10.1155/2008/306726
doi: 10.1155/2008/306726
Masuzaki R, Kanda T, Sasaki R, Matsumoto N, Ogawa M, Matsuoka S, Karp SJ, Moriyama M (2020) Noninvasive assessment of liver fibrosis: current and future clinical and molecular perspectives. Int J Mol Sci 21(14):4906. https://doi.org/10.3390/ijms21144906.
doi: 10.3390/ijms21144906.
pmcid: 7402287
Yoneda M, Honda Y, Nogami A, Imajo K, Nakajima A (2020) Advances in ultrasound elastography for nonalcoholic fatty liver disease. J Med Ultrason 47(4):521–533. https://doi.org/10.1007/s10396-020-01040-8
doi: 10.1007/s10396-020-01040-8