SCSONet: spatial-channel synergistic optimization net for skin lesion segmentation
In the field of computer-assisted medical diagnosis, developing medical image segmentation models that are both accurate and capable of real-time operation under limited computational resources is crucial. Particularly for skin disease image segmentation, the construction of such lightweight models...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2024-03-01
|
| Series: | Frontiers in Physics |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fphy.2024.1388364/full |
| _version_ | 1827313794732261376 |
|---|---|
| author | Haoyu Chen Zexin Li Xinyue Huang Zhengwei Peng Yichen Deng Li Tang Li Tang Li Yin Li Yin |
| author_facet | Haoyu Chen Zexin Li Xinyue Huang Zhengwei Peng Yichen Deng Li Tang Li Tang Li Yin Li Yin |
| author_sort | Haoyu Chen |
| collection | DOAJ |
| description | In the field of computer-assisted medical diagnosis, developing medical image segmentation models that are both accurate and capable of real-time operation under limited computational resources is crucial. Particularly for skin disease image segmentation, the construction of such lightweight models must balance computational cost and segmentation efficiency, especially in environments with limited computing power, memory, and storage. This study proposes a new lightweight network designed specifically for skin disease image segmentation, aimed at significantly reducing the number of parameters and floating-point operations while ensuring segmentation performance. The proposed ConvStem module, with full-dimensional attention, learns complementary attention weights across all four dimensions of the convolution kernel, effectively enhancing the recognition of irregularly shaped lesion areas, reducing the model’s parameter count and computational burden, thus promoting model lightweighting and performance improvement. The SCF Block reduces feature redundancy through spatial and channel feature fusion, significantly lowering parameter count while improving segmentation results. This paper validates the effectiveness and robustness of the proposed SCSONet on two public skin lesion segmentation datasets, demonstrating its low computational resource requirements. https://github.com/Haoyu1Chen/SCSONet. |
| first_indexed | 2024-04-24T22:22:21Z |
| format | Article |
| id | doaj.art-512083d7a12446c1964e96d9df343ff8 |
| institution | Directory Open Access Journal |
| issn | 2296-424X |
| language | English |
| last_indexed | 2024-04-24T22:22:21Z |
| publishDate | 2024-03-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Physics |
| spelling | doaj.art-512083d7a12446c1964e96d9df343ff82024-03-20T05:15:15ZengFrontiers Media S.A.Frontiers in Physics2296-424X2024-03-011210.3389/fphy.2024.13883641388364SCSONet: spatial-channel synergistic optimization net for skin lesion segmentationHaoyu Chen0Zexin Li1Xinyue Huang2Zhengwei Peng3Yichen Deng4Li Tang5Li Tang6Li Yin7Li Yin8College of Computer Science and Technology, Chongqing University of Posts and Telecommunications, Chongqing, ChinaCollege of Computer Science and Technology, Chongqing University of Posts and Telecommunications, Chongqing, ChinaCollege of Computer Science and Technology, Chongqing University of Posts and Telecommunications, Chongqing, ChinaCollege of Computer Science and Technology, Chongqing University of Posts and Telecommunications, Chongqing, ChinaCollege of Computer Science and Technology, Chongqing University of Posts and Telecommunications, Chongqing, ChinaDepartment of Radiation Oncology, Chongqing University Cancer Hospital, Chongqing, ChinaChongqing Key Laboratory of Translational Research for Cancer Metastasis and Individualized Treatment, Chongqing University Cancer Hospital, Chongqing, ChinaDepartment of Radiation Oncology, Chongqing University Cancer Hospital, Chongqing, ChinaChongqing Key Laboratory of Translational Research for Cancer Metastasis and Individualized Treatment, Chongqing University Cancer Hospital, Chongqing, ChinaIn the field of computer-assisted medical diagnosis, developing medical image segmentation models that are both accurate and capable of real-time operation under limited computational resources is crucial. Particularly for skin disease image segmentation, the construction of such lightweight models must balance computational cost and segmentation efficiency, especially in environments with limited computing power, memory, and storage. This study proposes a new lightweight network designed specifically for skin disease image segmentation, aimed at significantly reducing the number of parameters and floating-point operations while ensuring segmentation performance. The proposed ConvStem module, with full-dimensional attention, learns complementary attention weights across all four dimensions of the convolution kernel, effectively enhancing the recognition of irregularly shaped lesion areas, reducing the model’s parameter count and computational burden, thus promoting model lightweighting and performance improvement. The SCF Block reduces feature redundancy through spatial and channel feature fusion, significantly lowering parameter count while improving segmentation results. This paper validates the effectiveness and robustness of the proposed SCSONet on two public skin lesion segmentation datasets, demonstrating its low computational resource requirements. https://github.com/Haoyu1Chen/SCSONet.https://www.frontiersin.org/articles/10.3389/fphy.2024.1388364/fulllight-weight modelmedical image segmentationattention mechanismmobile healthskin lesion segmentation |
| spellingShingle | Haoyu Chen Zexin Li Xinyue Huang Zhengwei Peng Yichen Deng Li Tang Li Tang Li Yin Li Yin SCSONet: spatial-channel synergistic optimization net for skin lesion segmentation Frontiers in Physics light-weight model medical image segmentation attention mechanism mobile health skin lesion segmentation |
| title | SCSONet: spatial-channel synergistic optimization net for skin lesion segmentation |
| title_full | SCSONet: spatial-channel synergistic optimization net for skin lesion segmentation |
| title_fullStr | SCSONet: spatial-channel synergistic optimization net for skin lesion segmentation |
| title_full_unstemmed | SCSONet: spatial-channel synergistic optimization net for skin lesion segmentation |
| title_short | SCSONet: spatial-channel synergistic optimization net for skin lesion segmentation |
| title_sort | scsonet spatial channel synergistic optimization net for skin lesion segmentation |
| topic | light-weight model medical image segmentation attention mechanism mobile health skin lesion segmentation |
| url | https://www.frontiersin.org/articles/10.3389/fphy.2024.1388364/full |
| work_keys_str_mv | AT haoyuchen scsonetspatialchannelsynergisticoptimizationnetforskinlesionsegmentation AT zexinli scsonetspatialchannelsynergisticoptimizationnetforskinlesionsegmentation AT xinyuehuang scsonetspatialchannelsynergisticoptimizationnetforskinlesionsegmentation AT zhengweipeng scsonetspatialchannelsynergisticoptimizationnetforskinlesionsegmentation AT yichendeng scsonetspatialchannelsynergisticoptimizationnetforskinlesionsegmentation AT litang scsonetspatialchannelsynergisticoptimizationnetforskinlesionsegmentation AT litang scsonetspatialchannelsynergisticoptimizationnetforskinlesionsegmentation AT liyin scsonetspatialchannelsynergisticoptimizationnetforskinlesionsegmentation AT liyin scsonetspatialchannelsynergisticoptimizationnetforskinlesionsegmentation |