Abdominal multi-organ auto-segmentation using 3D-patch-based deep convolutional neural network

Abstract Segmentation of normal organs is a critical and time-consuming process in radiotherapy. Auto-segmentation of abdominal organs has been made possible by the advent of the convolutional neural network. We utilized the U-Net, a 3D-patch-based convolutional neural network, and added graph-cut a...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Hojin Kim, Jinhong Jung, Jieun Kim, Byungchul Cho, Jungwon Kwak, Jeong Yun Jang, Sang-wook Lee, June-Goo Lee, Sang Min Yoon
Formato:	article
Lenguaje:	EN
Publicado:	Nature Portfolio 2020
Materias:	Medicine R Science Q
Acceso en línea:	https://doaj.org/article/b15c3fcf9dfe41dbb039a77861818eb7
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:b15c3fcf9dfe41dbb039a77861818eb7
record_format	dspace
spelling	oai:doaj.org-article:b15c3fcf9dfe41dbb039a77861818eb72021-12-02T18:15:09ZAbdominal multi-organ auto-segmentation using 3D-patch-based deep convolutional neural network10.1038/s41598-020-63285-02045-2322https://doaj.org/article/b15c3fcf9dfe41dbb039a77861818eb72020-04-01T00:00:00Zhttps://doi.org/10.1038/s41598-020-63285-0https://doaj.org/toc/2045-2322Abstract Segmentation of normal organs is a critical and time-consuming process in radiotherapy. Auto-segmentation of abdominal organs has been made possible by the advent of the convolutional neural network. We utilized the U-Net, a 3D-patch-based convolutional neural network, and added graph-cut algorithm-based post-processing. The inputs were 3D-patch-based CT images consisting of 64 × 64 × 64 voxels designed to produce 3D multi-label semantic images representing the liver, stomach, duodenum, and right/left kidneys. The datasets for training, validating, and testing consisted of 80, 20, and 20 CT simulation scans, respectively. For accuracy assessment, the predicted structures were compared with those produced from the atlas-based method and inter-observer segmentation using the Dice similarity coefficient, Hausdorff distance, and mean surface distance. The efficiency was quantified by measuring the time elapsed for segmentation with or without automation using the U-Net. The U-Net-based auto-segmentation outperformed the atlas-based auto-segmentation in all abdominal structures, and showed comparable results to the inter-observer segmentations especially for liver and kidney. The average segmentation time without automation was 22.6 minutes, which was reduced to 7.1 minutes with automation using the U-Net. Our proposed auto-segmentation framework using the 3D-patch-based U-Net for abdominal multi-organs demonstrated potential clinical usefulness in terms of accuracy and time-efficiency.Hojin KimJinhong JungJieun KimByungchul ChoJungwon KwakJeong Yun JangSang-wook LeeJune-Goo LeeSang Min YoonNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 10, Iss 1, Pp 1-9 (2020)
institution	DOAJ
collection	DOAJ
language	EN
topic	Medicine R Science Q
spellingShingle	Medicine R Science Q Hojin Kim Jinhong Jung Jieun Kim Byungchul Cho Jungwon Kwak Jeong Yun Jang Sang-wook Lee June-Goo Lee Sang Min Yoon Abdominal multi-organ auto-segmentation using 3D-patch-based deep convolutional neural network
description	Abstract Segmentation of normal organs is a critical and time-consuming process in radiotherapy. Auto-segmentation of abdominal organs has been made possible by the advent of the convolutional neural network. We utilized the U-Net, a 3D-patch-based convolutional neural network, and added graph-cut algorithm-based post-processing. The inputs were 3D-patch-based CT images consisting of 64 × 64 × 64 voxels designed to produce 3D multi-label semantic images representing the liver, stomach, duodenum, and right/left kidneys. The datasets for training, validating, and testing consisted of 80, 20, and 20 CT simulation scans, respectively. For accuracy assessment, the predicted structures were compared with those produced from the atlas-based method and inter-observer segmentation using the Dice similarity coefficient, Hausdorff distance, and mean surface distance. The efficiency was quantified by measuring the time elapsed for segmentation with or without automation using the U-Net. The U-Net-based auto-segmentation outperformed the atlas-based auto-segmentation in all abdominal structures, and showed comparable results to the inter-observer segmentations especially for liver and kidney. The average segmentation time without automation was 22.6 minutes, which was reduced to 7.1 minutes with automation using the U-Net. Our proposed auto-segmentation framework using the 3D-patch-based U-Net for abdominal multi-organs demonstrated potential clinical usefulness in terms of accuracy and time-efficiency.
format	article
author	Hojin Kim Jinhong Jung Jieun Kim Byungchul Cho Jungwon Kwak Jeong Yun Jang Sang-wook Lee June-Goo Lee Sang Min Yoon
author_facet	Hojin Kim Jinhong Jung Jieun Kim Byungchul Cho Jungwon Kwak Jeong Yun Jang Sang-wook Lee June-Goo Lee Sang Min Yoon
author_sort	Hojin Kim
title	Abdominal multi-organ auto-segmentation using 3D-patch-based deep convolutional neural network
title_short	Abdominal multi-organ auto-segmentation using 3D-patch-based deep convolutional neural network
title_full	Abdominal multi-organ auto-segmentation using 3D-patch-based deep convolutional neural network
title_fullStr	Abdominal multi-organ auto-segmentation using 3D-patch-based deep convolutional neural network
title_full_unstemmed	Abdominal multi-organ auto-segmentation using 3D-patch-based deep convolutional neural network
title_sort	abdominal multi-organ auto-segmentation using 3d-patch-based deep convolutional neural network
publisher	Nature Portfolio
publishDate	2020
url	https://doaj.org/article/b15c3fcf9dfe41dbb039a77861818eb7
work_keys_str_mv	AT hojinkim abdominalmultiorganautosegmentationusing3dpatchbaseddeepconvolutionalneuralnetwork AT jinhongjung abdominalmultiorganautosegmentationusing3dpatchbaseddeepconvolutionalneuralnetwork AT jieunkim abdominalmultiorganautosegmentationusing3dpatchbaseddeepconvolutionalneuralnetwork AT byungchulcho abdominalmultiorganautosegmentationusing3dpatchbaseddeepconvolutionalneuralnetwork AT jungwonkwak abdominalmultiorganautosegmentationusing3dpatchbaseddeepconvolutionalneuralnetwork AT jeongyunjang abdominalmultiorganautosegmentationusing3dpatchbaseddeepconvolutionalneuralnetwork AT sangwooklee abdominalmultiorganautosegmentationusing3dpatchbaseddeepconvolutionalneuralnetwork AT junegoolee abdominalmultiorganautosegmentationusing3dpatchbaseddeepconvolutionalneuralnetwork AT sangminyoon abdominalmultiorganautosegmentationusing3dpatchbaseddeepconvolutionalneuralnetwork
_version_	1718378415205122048

Abdominal multi-organ auto-segmentation using 3D-patch-based deep convolutional neural network

Ejemplares similares