Brain Tumor Segmentation Network with Multi-View Ensemble Discrimination and Kernel-Sharing Dilated Convolution
Accurate segmentation of brain tumors from magnetic resonance 3D images (MRI) is critical for clinical decisions and surgical planning. Radiologists usually separate and analyze brain tumors by combining images of axial, coronal, and sagittal views. However, traditional convolutional neural network...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-04-01
|
| Series: | Brain Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3425/13/4/650 |
| _version_ | 1797606118743080960 |
|---|---|
| author | Xin Guan Yushan Zhao Charles Okanda Nyatega Qiang Li |
| author_facet | Xin Guan Yushan Zhao Charles Okanda Nyatega Qiang Li |
| author_sort | Xin Guan |
| collection | DOAJ |
| description | Accurate segmentation of brain tumors from magnetic resonance 3D images (MRI) is critical for clinical decisions and surgical planning. Radiologists usually separate and analyze brain tumors by combining images of axial, coronal, and sagittal views. However, traditional convolutional neural network (CNN) models tend to use information from only a single view or one by one. Moreover, the existing models adopt a multi-branch structure with different-size convolution kernels in parallel to adapt to various tumor sizes. However, the difference in the convolution kernels’ parameters cannot precisely characterize the feature similarity of tumor lesion regions with various sizes, connectivity, and convexity. To address the above problems, we propose a hierarchical multi-view convolution method that decouples the standard 3D convolution into axial, coronal, and sagittal views to provide complementary-view features. Then, every pixel is classified by ensembling the discriminant results from the three views. Moreover, we propose a multi-branch kernel-sharing mechanism with a dilated rate to obtain parameter-consistent convolution kernels with different receptive fields. We use the BraTS2018 and BraTS2020 datasets for comparison experiments. The average Dice coefficients of the proposed network on the BraTS2020 dataset can reach 78.16%, 89.52%, and 83.05% for the enhancing tumor (ET), whole tumor (WT), and tumor core (TC), respectively, while the number of parameters is only 0.5 M. Compared with the baseline network for brain tumor segmentation, the accuracy was improved by 1.74%, 0.5%, and 2.19%, respectively. |
| first_indexed | 2024-03-11T05:10:47Z |
| format | Article |
| id | doaj.art-24ce1b9489f547d6a5641cce549056d7 |
| institution | Directory Open Access Journal |
| issn | 2076-3425 |
| language | English |
| last_indexed | 2024-03-11T05:10:47Z |
| publishDate | 2023-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Brain Sciences |
| spelling | doaj.art-24ce1b9489f547d6a5641cce549056d72023-11-17T18:33:08ZengMDPI AGBrain Sciences2076-34252023-04-0113465010.3390/brainsci13040650Brain Tumor Segmentation Network with Multi-View Ensemble Discrimination and Kernel-Sharing Dilated ConvolutionXin Guan0Yushan Zhao1Charles Okanda Nyatega2Qiang Li3School of Microelectronics, Tianjin University, Tianjin 300072, ChinaSchool of Microelectronics, Tianjin University, Tianjin 300072, ChinaSchool of Electrical and Information Engineering, Tianjin University, Tianjin 300072, ChinaSchool of Microelectronics, Tianjin University, Tianjin 300072, ChinaAccurate segmentation of brain tumors from magnetic resonance 3D images (MRI) is critical for clinical decisions and surgical planning. Radiologists usually separate and analyze brain tumors by combining images of axial, coronal, and sagittal views. However, traditional convolutional neural network (CNN) models tend to use information from only a single view or one by one. Moreover, the existing models adopt a multi-branch structure with different-size convolution kernels in parallel to adapt to various tumor sizes. However, the difference in the convolution kernels’ parameters cannot precisely characterize the feature similarity of tumor lesion regions with various sizes, connectivity, and convexity. To address the above problems, we propose a hierarchical multi-view convolution method that decouples the standard 3D convolution into axial, coronal, and sagittal views to provide complementary-view features. Then, every pixel is classified by ensembling the discriminant results from the three views. Moreover, we propose a multi-branch kernel-sharing mechanism with a dilated rate to obtain parameter-consistent convolution kernels with different receptive fields. We use the BraTS2018 and BraTS2020 datasets for comparison experiments. The average Dice coefficients of the proposed network on the BraTS2020 dataset can reach 78.16%, 89.52%, and 83.05% for the enhancing tumor (ET), whole tumor (WT), and tumor core (TC), respectively, while the number of parameters is only 0.5 M. Compared with the baseline network for brain tumor segmentation, the accuracy was improved by 1.74%, 0.5%, and 2.19%, respectively.https://www.mdpi.com/2076-3425/13/4/650brain tumor segmentationhierarchical multi-view convolutionensemble discriminationfeature similarityvarious sizesdeep learning |
| spellingShingle | Xin Guan Yushan Zhao Charles Okanda Nyatega Qiang Li Brain Tumor Segmentation Network with Multi-View Ensemble Discrimination and Kernel-Sharing Dilated Convolution Brain Sciences brain tumor segmentation hierarchical multi-view convolution ensemble discrimination feature similarity various sizes deep learning |
| title | Brain Tumor Segmentation Network with Multi-View Ensemble Discrimination and Kernel-Sharing Dilated Convolution |
| title_full | Brain Tumor Segmentation Network with Multi-View Ensemble Discrimination and Kernel-Sharing Dilated Convolution |
| title_fullStr | Brain Tumor Segmentation Network with Multi-View Ensemble Discrimination and Kernel-Sharing Dilated Convolution |
| title_full_unstemmed | Brain Tumor Segmentation Network with Multi-View Ensemble Discrimination and Kernel-Sharing Dilated Convolution |
| title_short | Brain Tumor Segmentation Network with Multi-View Ensemble Discrimination and Kernel-Sharing Dilated Convolution |
| title_sort | brain tumor segmentation network with multi view ensemble discrimination and kernel sharing dilated convolution |
| topic | brain tumor segmentation hierarchical multi-view convolution ensemble discrimination feature similarity various sizes deep learning |
| url | https://www.mdpi.com/2076-3425/13/4/650 |
| work_keys_str_mv | AT xinguan braintumorsegmentationnetworkwithmultiviewensemblediscriminationandkernelsharingdilatedconvolution AT yushanzhao braintumorsegmentationnetworkwithmultiviewensemblediscriminationandkernelsharingdilatedconvolution AT charlesokandanyatega braintumorsegmentationnetworkwithmultiviewensemblediscriminationandkernelsharingdilatedconvolution AT qiangli braintumorsegmentationnetworkwithmultiviewensemblediscriminationandkernelsharingdilatedconvolution |