Automatic Segmentation of Organs-at-Risk from Head-and-Neck CT using Separable Convolutional Neural Network with Hard-Region-Weighted Loss
Nasopharyngeal Carcinoma (NPC) is a leading form of Head-and-Neck (HAN) cancer in the Arctic, China, Southeast Asia, and the Middle East/North Africa. Accurate segmentation of Organs-at-Risk (OAR) from Computed Tomography (CT) images with uncertainty information is critical for effective planning of...
Main Authors: | , , , , , , , , , |
---|---|
Format: | Article in Journal/Newspaper |
Language: | unknown |
Published: |
arXiv
2021
|
Subjects: | |
Online Access: | https://dx.doi.org/10.48550/arxiv.2102.01897 https://arxiv.org/abs/2102.01897 |
id |
ftdatacite:10.48550/arxiv.2102.01897 |
---|---|
record_format |
openpolar |
spelling |
ftdatacite:10.48550/arxiv.2102.01897 2023-05-15T15:11:58+02:00 Automatic Segmentation of Organs-at-Risk from Head-and-Neck CT using Separable Convolutional Neural Network with Hard-Region-Weighted Loss Lei, Wenhui Mei, Haochen Sun, Zhengwentai Ye, Shan Gu, Ran Wang, Huan Huang, Rui Zhang, Shichuan Zhang, Shaoting Wang, Guotai 2021 https://dx.doi.org/10.48550/arxiv.2102.01897 https://arxiv.org/abs/2102.01897 unknown arXiv Creative Commons Zero v1.0 Universal https://creativecommons.org/publicdomain/zero/1.0/legalcode cc0-1.0 CC0 Image and Video Processing eess.IV Computer Vision and Pattern Recognition cs.CV FOS Electrical engineering, electronic engineering, information engineering FOS Computer and information sciences Article CreativeWork article Preprint 2021 ftdatacite https://doi.org/10.48550/arxiv.2102.01897 2022-03-10T14:47:09Z Nasopharyngeal Carcinoma (NPC) is a leading form of Head-and-Neck (HAN) cancer in the Arctic, China, Southeast Asia, and the Middle East/North Africa. Accurate segmentation of Organs-at-Risk (OAR) from Computed Tomography (CT) images with uncertainty information is critical for effective planning of radiation therapy for NPC treatment. Despite the stateof-the-art performance achieved by Convolutional Neural Networks (CNNs) for automatic segmentation of OARs, existing methods do not provide uncertainty estimation of the segmentation results for treatment planning, and their accuracy is still limited by several factors, including the low contrast of soft tissues in CT, highly imbalanced sizes of OARs and large inter-slice spacing. To address these problems, we propose a novel framework for accurate OAR segmentation with reliable uncertainty estimation. First, we propose a Segmental Linear Function (SLF) to transform the intensity of CT images to make multiple organs more distinguishable than existing methods based on a simple window width/level that often gives a better visibility of one organ while hiding the others. Second, to deal with the large inter-slice spacing, we introduce a novel 2.5D network (named as 3D-SepNet) specially designed for dealing with clinic HAN CT scans with anisotropic spacing. Thirdly, existing hardness-aware loss function often deal with class-level hardness, but our proposed attention to hard voxels (ATH) uses a voxel-level hardness strategy, which is more suitable to dealing with some hard regions despite that its corresponding class may be easy. Our code is now available at https://github.com/HiLab-git/SepNet. : Accepted by Neurocomputing Article in Journal/Newspaper Arctic DataCite Metadata Store (German National Library of Science and Technology) Arctic |
institution |
Open Polar |
collection |
DataCite Metadata Store (German National Library of Science and Technology) |
op_collection_id |
ftdatacite |
language |
unknown |
topic |
Image and Video Processing eess.IV Computer Vision and Pattern Recognition cs.CV FOS Electrical engineering, electronic engineering, information engineering FOS Computer and information sciences |
spellingShingle |
Image and Video Processing eess.IV Computer Vision and Pattern Recognition cs.CV FOS Electrical engineering, electronic engineering, information engineering FOS Computer and information sciences Lei, Wenhui Mei, Haochen Sun, Zhengwentai Ye, Shan Gu, Ran Wang, Huan Huang, Rui Zhang, Shichuan Zhang, Shaoting Wang, Guotai Automatic Segmentation of Organs-at-Risk from Head-and-Neck CT using Separable Convolutional Neural Network with Hard-Region-Weighted Loss |
topic_facet |
Image and Video Processing eess.IV Computer Vision and Pattern Recognition cs.CV FOS Electrical engineering, electronic engineering, information engineering FOS Computer and information sciences |
description |
Nasopharyngeal Carcinoma (NPC) is a leading form of Head-and-Neck (HAN) cancer in the Arctic, China, Southeast Asia, and the Middle East/North Africa. Accurate segmentation of Organs-at-Risk (OAR) from Computed Tomography (CT) images with uncertainty information is critical for effective planning of radiation therapy for NPC treatment. Despite the stateof-the-art performance achieved by Convolutional Neural Networks (CNNs) for automatic segmentation of OARs, existing methods do not provide uncertainty estimation of the segmentation results for treatment planning, and their accuracy is still limited by several factors, including the low contrast of soft tissues in CT, highly imbalanced sizes of OARs and large inter-slice spacing. To address these problems, we propose a novel framework for accurate OAR segmentation with reliable uncertainty estimation. First, we propose a Segmental Linear Function (SLF) to transform the intensity of CT images to make multiple organs more distinguishable than existing methods based on a simple window width/level that often gives a better visibility of one organ while hiding the others. Second, to deal with the large inter-slice spacing, we introduce a novel 2.5D network (named as 3D-SepNet) specially designed for dealing with clinic HAN CT scans with anisotropic spacing. Thirdly, existing hardness-aware loss function often deal with class-level hardness, but our proposed attention to hard voxels (ATH) uses a voxel-level hardness strategy, which is more suitable to dealing with some hard regions despite that its corresponding class may be easy. Our code is now available at https://github.com/HiLab-git/SepNet. : Accepted by Neurocomputing |
format |
Article in Journal/Newspaper |
author |
Lei, Wenhui Mei, Haochen Sun, Zhengwentai Ye, Shan Gu, Ran Wang, Huan Huang, Rui Zhang, Shichuan Zhang, Shaoting Wang, Guotai |
author_facet |
Lei, Wenhui Mei, Haochen Sun, Zhengwentai Ye, Shan Gu, Ran Wang, Huan Huang, Rui Zhang, Shichuan Zhang, Shaoting Wang, Guotai |
author_sort |
Lei, Wenhui |
title |
Automatic Segmentation of Organs-at-Risk from Head-and-Neck CT using Separable Convolutional Neural Network with Hard-Region-Weighted Loss |
title_short |
Automatic Segmentation of Organs-at-Risk from Head-and-Neck CT using Separable Convolutional Neural Network with Hard-Region-Weighted Loss |
title_full |
Automatic Segmentation of Organs-at-Risk from Head-and-Neck CT using Separable Convolutional Neural Network with Hard-Region-Weighted Loss |
title_fullStr |
Automatic Segmentation of Organs-at-Risk from Head-and-Neck CT using Separable Convolutional Neural Network with Hard-Region-Weighted Loss |
title_full_unstemmed |
Automatic Segmentation of Organs-at-Risk from Head-and-Neck CT using Separable Convolutional Neural Network with Hard-Region-Weighted Loss |
title_sort |
automatic segmentation of organs-at-risk from head-and-neck ct using separable convolutional neural network with hard-region-weighted loss |
publisher |
arXiv |
publishDate |
2021 |
url |
https://dx.doi.org/10.48550/arxiv.2102.01897 https://arxiv.org/abs/2102.01897 |
geographic |
Arctic |
geographic_facet |
Arctic |
genre |
Arctic |
genre_facet |
Arctic |
op_rights |
Creative Commons Zero v1.0 Universal https://creativecommons.org/publicdomain/zero/1.0/legalcode cc0-1.0 |
op_rightsnorm |
CC0 |
op_doi |
https://doi.org/10.48550/arxiv.2102.01897 |
_version_ |
1766342734623277056 |