MSDF-Net: Multi-Scale Deep Fusion Network for Stroke Lesion Segmentation
Lesion segmentation is of great research interest due to its capability in facilitating accurate stroke diagnosis and surgical planning. Existing deep neural networks, such as U-net, have demonstrated encouraging progress in biomedical image segmentation. Nevertheless, there are still many challenge...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2019-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/8928498/ |
| _version_ | 1818877613209288704 |
|---|---|
| author | Xinfeng Liu Hao Yang Kehan Qi Pei Dong Qiegen Liu Xin Liu Rongpin Wang Shanshan Wang |
| author_facet | Xinfeng Liu Hao Yang Kehan Qi Pei Dong Qiegen Liu Xin Liu Rongpin Wang Shanshan Wang |
| author_sort | Xinfeng Liu |
| collection | DOAJ |
| description | Lesion segmentation is of great research interest due to its capability in facilitating accurate stroke diagnosis and surgical planning. Existing deep neural networks, such as U-net, have demonstrated encouraging progress in biomedical image segmentation. Nevertheless, there are still many challenges related to the segmentation of stroke lesions, including dealing with diverse lesion locations, variations in lesion scales, and fuzzy lesion boundaries. In order to address these challenges, this paper proposes a deep neural network architecture denoted as the Multi-Scale Deep Fusion Network (MSDF-Net) with Atrous Spatial Pyramid Pooling (ASPP) for the feature extraction at different scales, and the inclusion of capsules to deal with complicated relative entities. The proposed method is essentially an end-to-end deep encoder-decoder neural network. The cross connection between the encoder and the decoder guarantees the high resolution of the feature mapping. Experimental results on the open-source Anatomical Tracings of Lesions After Stroke (ATLAS) dataset shows that the proposed model achieved a higher evaluating score compared to 5 existing models. |
| first_indexed | 2024-12-19T14:01:04Z |
| format | Article |
| id | doaj.art-62f9dc370305405cb30b7c953d702950 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-19T14:01:04Z |
| publishDate | 2019-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-62f9dc370305405cb30b7c953d7029502022-12-21T20:18:27ZengIEEEIEEE Access2169-35362019-01-01717848617849510.1109/ACCESS.2019.29583848928498MSDF-Net: Multi-Scale Deep Fusion Network for Stroke Lesion SegmentationXinfeng Liu0https://orcid.org/0000-0001-9075-7207Hao Yang1https://orcid.org/0000-0002-5616-7796Kehan Qi2https://orcid.org/0000-0002-0575-6523Pei Dong3https://orcid.org/0000-0002-3122-3739Qiegen Liu4https://orcid.org/0000-0003-4717-2283Xin Liu5Rongpin Wang6Shanshan Wang7Guizhou Provincial People’s Hospital, Guiyang, ChinaShenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen, ChinaShenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen, ChinaTencent Corporation, Shenzhen, ChinaDepartment of Electronic Information Engineering, Nanchang University, Nanchang, ChinaShenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen, ChinaGuizhou Provincial People’s Hospital, Guiyang, ChinaShenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen, ChinaLesion segmentation is of great research interest due to its capability in facilitating accurate stroke diagnosis and surgical planning. Existing deep neural networks, such as U-net, have demonstrated encouraging progress in biomedical image segmentation. Nevertheless, there are still many challenges related to the segmentation of stroke lesions, including dealing with diverse lesion locations, variations in lesion scales, and fuzzy lesion boundaries. In order to address these challenges, this paper proposes a deep neural network architecture denoted as the Multi-Scale Deep Fusion Network (MSDF-Net) with Atrous Spatial Pyramid Pooling (ASPP) for the feature extraction at different scales, and the inclusion of capsules to deal with complicated relative entities. The proposed method is essentially an end-to-end deep encoder-decoder neural network. The cross connection between the encoder and the decoder guarantees the high resolution of the feature mapping. Experimental results on the open-source Anatomical Tracings of Lesions After Stroke (ATLAS) dataset shows that the proposed model achieved a higher evaluating score compared to 5 existing models.https://ieeexplore.ieee.org/document/8928498/Lesion segmentationdeep neural networkatrous spatial pyramid poolingcapsule networkdynamic routing |
| spellingShingle | Xinfeng Liu Hao Yang Kehan Qi Pei Dong Qiegen Liu Xin Liu Rongpin Wang Shanshan Wang MSDF-Net: Multi-Scale Deep Fusion Network for Stroke Lesion Segmentation IEEE Access Lesion segmentation deep neural network atrous spatial pyramid pooling capsule network dynamic routing |
| title | MSDF-Net: Multi-Scale Deep Fusion Network for Stroke Lesion Segmentation |
| title_full | MSDF-Net: Multi-Scale Deep Fusion Network for Stroke Lesion Segmentation |
| title_fullStr | MSDF-Net: Multi-Scale Deep Fusion Network for Stroke Lesion Segmentation |
| title_full_unstemmed | MSDF-Net: Multi-Scale Deep Fusion Network for Stroke Lesion Segmentation |
| title_short | MSDF-Net: Multi-Scale Deep Fusion Network for Stroke Lesion Segmentation |
| title_sort | msdf net multi scale deep fusion network for stroke lesion segmentation |
| topic | Lesion segmentation deep neural network atrous spatial pyramid pooling capsule network dynamic routing |
| url | https://ieeexplore.ieee.org/document/8928498/ |
| work_keys_str_mv | AT xinfengliu msdfnetmultiscaledeepfusionnetworkforstrokelesionsegmentation AT haoyang msdfnetmultiscaledeepfusionnetworkforstrokelesionsegmentation AT kehanqi msdfnetmultiscaledeepfusionnetworkforstrokelesionsegmentation AT peidong msdfnetmultiscaledeepfusionnetworkforstrokelesionsegmentation AT qiegenliu msdfnetmultiscaledeepfusionnetworkforstrokelesionsegmentation AT xinliu msdfnetmultiscaledeepfusionnetworkforstrokelesionsegmentation AT rongpinwang msdfnetmultiscaledeepfusionnetworkforstrokelesionsegmentation AT shanshanwang msdfnetmultiscaledeepfusionnetworkforstrokelesionsegmentation |