Automatic Pancreas Segmentation via Coarse Location and Ensemble Learning
Automatic and reliable segmentation of the pancreas is an important but difficult task for various clinical applications, such as pancreatic cancer radiotherapy and computer-aided diagnosis (CAD). The main challenges for accurate CT pancreas segmentation lie in two aspects: (1) large shape variation...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2020-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/8937496/ |
| _version_ | 1818665657348128768 |
|---|---|
| author | Shangqing Liu Xinrui Yuan Runyue Hu Shujun Liang Shaohua Feng Yuhua Ai Yu Zhang |
| author_facet | Shangqing Liu Xinrui Yuan Runyue Hu Shujun Liang Shaohua Feng Yuhua Ai Yu Zhang |
| author_sort | Shangqing Liu |
| collection | DOAJ |
| description | Automatic and reliable segmentation of the pancreas is an important but difficult task for various clinical applications, such as pancreatic cancer radiotherapy and computer-aided diagnosis (CAD). The main challenges for accurate CT pancreas segmentation lie in two aspects: (1) large shape variation across different patients, and (2) low contrast and blurring around the pancreas boundary. In this paper, we propose a two-stage, ensemble-based fully convolutional neural network (FCN) to solve the challenging pancreas segmentation problem in CT images. First, candidate region generation is performed by classifying patches generated by superpixels. Second, five FCNs based on the U-Net architecture are trained with different objective functions. For each network, 2.5D slices are used as the input to provide 3D image information complementarily without the need for computationally expensive 3D convolutions. Then, an ensemble model is utilized to combine the five output segmentation maps and achieve the final segmentation. The proposed method is extensively evaluated on a publicly available dataset of 82 manually segmented CT volumes via 4-fold cross-validation. Experimental results show its superior performance compared with several state-of-the-art methods with a Dice coefficient of 84.10±4.91% and Jaccard coefficient of 72.86±6.89%. |
| first_indexed | 2024-12-17T05:52:07Z |
| format | Article |
| id | doaj.art-fe6453e868984903b6a0ead7b222c7f0 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-17T05:52:07Z |
| publishDate | 2020-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-fe6453e868984903b6a0ead7b222c7f02022-12-21T22:01:08ZengIEEEIEEE Access2169-35362020-01-0182906291410.1109/ACCESS.2019.29611258937496Automatic Pancreas Segmentation via Coarse Location and Ensemble LearningShangqing Liu0https://orcid.org/0000-0003-1883-8090Xinrui Yuan1Runyue Hu2Shujun Liang3Shaohua Feng4Yuhua Ai5Yu Zhang6School of Biomedical Engineering, Southern Medical University, Guangzhou, ChinaSchool of Biomedical Engineering, Southern Medical University, Guangzhou, ChinaSchool of Biomedical Engineering, Southern Medical University, Guangzhou, ChinaSchool of Biomedical Engineering, Southern Medical University, Guangzhou, ChinaRadiology Department, Manzhouli People’s Hospital, Manzhouli, ChinaSchool of Biomedical Engineering, Southern Medical University, Guangzhou, ChinaSchool of Biomedical Engineering, Southern Medical University, Guangzhou, ChinaAutomatic and reliable segmentation of the pancreas is an important but difficult task for various clinical applications, such as pancreatic cancer radiotherapy and computer-aided diagnosis (CAD). The main challenges for accurate CT pancreas segmentation lie in two aspects: (1) large shape variation across different patients, and (2) low contrast and blurring around the pancreas boundary. In this paper, we propose a two-stage, ensemble-based fully convolutional neural network (FCN) to solve the challenging pancreas segmentation problem in CT images. First, candidate region generation is performed by classifying patches generated by superpixels. Second, five FCNs based on the U-Net architecture are trained with different objective functions. For each network, 2.5D slices are used as the input to provide 3D image information complementarily without the need for computationally expensive 3D convolutions. Then, an ensemble model is utilized to combine the five output segmentation maps and achieve the final segmentation. The proposed method is extensively evaluated on a publicly available dataset of 82 manually segmented CT volumes via 4-fold cross-validation. Experimental results show its superior performance compared with several state-of-the-art methods with a Dice coefficient of 84.10±4.91% and Jaccard coefficient of 72.86±6.89%.https://ieeexplore.ieee.org/document/8937496/SuperpixelResNetfully convolutional neural networksensemble learningpancreas segmentation |
| spellingShingle | Shangqing Liu Xinrui Yuan Runyue Hu Shujun Liang Shaohua Feng Yuhua Ai Yu Zhang Automatic Pancreas Segmentation via Coarse Location and Ensemble Learning IEEE Access Superpixel ResNet fully convolutional neural networks ensemble learning pancreas segmentation |
| title | Automatic Pancreas Segmentation via Coarse Location and Ensemble Learning |
| title_full | Automatic Pancreas Segmentation via Coarse Location and Ensemble Learning |
| title_fullStr | Automatic Pancreas Segmentation via Coarse Location and Ensemble Learning |
| title_full_unstemmed | Automatic Pancreas Segmentation via Coarse Location and Ensemble Learning |
| title_short | Automatic Pancreas Segmentation via Coarse Location and Ensemble Learning |
| title_sort | automatic pancreas segmentation via coarse location and ensemble learning |
| topic | Superpixel ResNet fully convolutional neural networks ensemble learning pancreas segmentation |
| url | https://ieeexplore.ieee.org/document/8937496/ |
| work_keys_str_mv | AT shangqingliu automaticpancreassegmentationviacoarselocationandensemblelearning AT xinruiyuan automaticpancreassegmentationviacoarselocationandensemblelearning AT runyuehu automaticpancreassegmentationviacoarselocationandensemblelearning AT shujunliang automaticpancreassegmentationviacoarselocationandensemblelearning AT shaohuafeng automaticpancreassegmentationviacoarselocationandensemblelearning AT yuhuaai automaticpancreassegmentationviacoarselocationandensemblelearning AT yuzhang automaticpancreassegmentationviacoarselocationandensemblelearning |