A mutual inclusion mechanism for precise boundary segmentation in medical images
IntroductionAccurate image segmentation is crucial in medical imaging for quantifying diseases, assessing prognosis, and evaluating treatment outcomes. However, existing methods often fall short in integrating global and local features in a meaningful way, failing to give sufficient attention to abn...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2024-12-01
|
| Series: | Frontiers in Bioengineering and Biotechnology |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fbioe.2024.1504249/full |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1846110301832871936 |
|---|---|
| author | Yizhi Pan Junyi Xin Junyi Xin Tianhua Yang Siqi Li Le-Minh Nguyen Teeradaj Racharak Kai Li Kai Li Guanqun Sun Guanqun Sun Guanqun Sun |
| author_facet | Yizhi Pan Junyi Xin Junyi Xin Tianhua Yang Siqi Li Le-Minh Nguyen Teeradaj Racharak Kai Li Kai Li Guanqun Sun Guanqun Sun Guanqun Sun |
| author_sort | Yizhi Pan |
| collection | DOAJ |
| description | IntroductionAccurate image segmentation is crucial in medical imaging for quantifying diseases, assessing prognosis, and evaluating treatment outcomes. However, existing methods often fall short in integrating global and local features in a meaningful way, failing to give sufficient attention to abnormal regions and boundary details in medical images. These limitations hinder the effectiveness of segmentation techniques in clinical settings. To address these issues, we propose a novel deep learning-based approach, MIPC-Net, designed for precise boundary segmentation in medical images.MethodsOur approach, inspired by radiologists' working patterns, introduces two distinct modules: 1. Mutual Inclusion of Position and Channel Attention (MIPC) Module: To improve boundary segmentation precision, we present the MIPC module. This module enhances the focus on channel information while extracting position features and vice versa, effectively enhancing the segmentation of boundaries in medical images. 2. Skip-Residue Module: To optimize the restoration of medical images, we introduce Skip-Residue, a global residual connection. This module improves the integration of the encoder and decoder by filtering out irrelevant information and recovering the most crucial information lost during the feature extraction process.ResultsWe evaluate the performance of MIPC-Net on three publicly accessible datasets: Synapse, ISIC2018-Task, and Segpc. The evaluation uses metrics such as the Dice coefficient (DSC) and Hausdorff Distance (HD). Our ablation study confirms that each module contributes to the overall improvement of segmentation quality. Notably, with the integration of both modules, our model outperforms state-of-the-art methods across all metrics. Specifically, MIPC-Net achieves a 2.23 mm reduction in Hausdorff Distance on the Synapse dataset, highlighting the model's enhanced capability for precise image boundary segmentation.ConclusionThe introduction of the novel MIPC and Skip-Residue modules significantly improves feature extraction accuracy, leading to better boundary recognition in medical image segmentation tasks. Our approach demonstrates substantial improvements over existing methods, as evidenced by the results on benchmark datasets. |
| format | Article |
| id | doaj-art-14ca60ed2e2d4fd9b8d5d9e8cb1618ee |
| institution | Kabale University |
| issn | 2296-4185 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Bioengineering and Biotechnology |
| spelling | doaj-art-14ca60ed2e2d4fd9b8d5d9e8cb1618ee2024-12-24T09:36:47ZengFrontiers Media S.A.Frontiers in Bioengineering and Biotechnology2296-41852024-12-011210.3389/fbioe.2024.15042491504249A mutual inclusion mechanism for precise boundary segmentation in medical imagesYizhi Pan0Junyi Xin1Junyi Xin2Tianhua Yang3Siqi Li4Le-Minh Nguyen5Teeradaj Racharak6Kai Li7Kai Li8Guanqun Sun9Guanqun Sun10Guanqun Sun11School of Information Engineering, Zhejiang Provincial People's Hospital, Affiliated People's Hospital, Hangzhou Medical College, Hangzhou, ChinaSchool of Information Engineering, Zhejiang Provincial People's Hospital, Affiliated People's Hospital, Hangzhou Medical College, Hangzhou, ChinaZhejiang Provincial Engineering Research Center for Brain Cognition, Disease and Digital Medical Devices, Hangzhou Medical College, Hangzhou, ChinaSchool of Information Engineering, Zhejiang Provincial People's Hospital, Affiliated People's Hospital, Hangzhou Medical College, Hangzhou, ChinaSchool of Information Science, Japan Advanced Institute of Science and Technology, Nomi, Ishikawa, JapanSchool of Information Science, Japan Advanced Institute of Science and Technology, Nomi, Ishikawa, JapanSchool of Information Science, Japan Advanced Institute of Science and Technology, Nomi, Ishikawa, JapanSchool of Information Engineering, Zhejiang Provincial People's Hospital, Affiliated People's Hospital, Hangzhou Medical College, Hangzhou, ChinaZhejiang Provincial Engineering Research Center for Brain Cognition, Disease and Digital Medical Devices, Hangzhou Medical College, Hangzhou, ChinaSchool of Information Engineering, Zhejiang Provincial People's Hospital, Affiliated People's Hospital, Hangzhou Medical College, Hangzhou, ChinaZhejiang Provincial Engineering Research Center for Brain Cognition, Disease and Digital Medical Devices, Hangzhou Medical College, Hangzhou, ChinaSchool of Information Science, Japan Advanced Institute of Science and Technology, Nomi, Ishikawa, JapanIntroductionAccurate image segmentation is crucial in medical imaging for quantifying diseases, assessing prognosis, and evaluating treatment outcomes. However, existing methods often fall short in integrating global and local features in a meaningful way, failing to give sufficient attention to abnormal regions and boundary details in medical images. These limitations hinder the effectiveness of segmentation techniques in clinical settings. To address these issues, we propose a novel deep learning-based approach, MIPC-Net, designed for precise boundary segmentation in medical images.MethodsOur approach, inspired by radiologists' working patterns, introduces two distinct modules: 1. Mutual Inclusion of Position and Channel Attention (MIPC) Module: To improve boundary segmentation precision, we present the MIPC module. This module enhances the focus on channel information while extracting position features and vice versa, effectively enhancing the segmentation of boundaries in medical images. 2. Skip-Residue Module: To optimize the restoration of medical images, we introduce Skip-Residue, a global residual connection. This module improves the integration of the encoder and decoder by filtering out irrelevant information and recovering the most crucial information lost during the feature extraction process.ResultsWe evaluate the performance of MIPC-Net on three publicly accessible datasets: Synapse, ISIC2018-Task, and Segpc. The evaluation uses metrics such as the Dice coefficient (DSC) and Hausdorff Distance (HD). Our ablation study confirms that each module contributes to the overall improvement of segmentation quality. Notably, with the integration of both modules, our model outperforms state-of-the-art methods across all metrics. Specifically, MIPC-Net achieves a 2.23 mm reduction in Hausdorff Distance on the Synapse dataset, highlighting the model's enhanced capability for precise image boundary segmentation.ConclusionThe introduction of the novel MIPC and Skip-Residue modules significantly improves feature extraction accuracy, leading to better boundary recognition in medical image segmentation tasks. Our approach demonstrates substantial improvements over existing methods, as evidenced by the results on benchmark datasets.https://www.frontiersin.org/articles/10.3389/fbioe.2024.1504249/fullU-Netmedical image segmentationmutual inclusiontransformerdeep learning |
| spellingShingle | Yizhi Pan Junyi Xin Junyi Xin Tianhua Yang Siqi Li Le-Minh Nguyen Teeradaj Racharak Kai Li Kai Li Guanqun Sun Guanqun Sun Guanqun Sun A mutual inclusion mechanism for precise boundary segmentation in medical images Frontiers in Bioengineering and Biotechnology U-Net medical image segmentation mutual inclusion transformer deep learning |
| title | A mutual inclusion mechanism for precise boundary segmentation in medical images |
| title_full | A mutual inclusion mechanism for precise boundary segmentation in medical images |
| title_fullStr | A mutual inclusion mechanism for precise boundary segmentation in medical images |
| title_full_unstemmed | A mutual inclusion mechanism for precise boundary segmentation in medical images |
| title_short | A mutual inclusion mechanism for precise boundary segmentation in medical images |
| title_sort | mutual inclusion mechanism for precise boundary segmentation in medical images |
| topic | U-Net medical image segmentation mutual inclusion transformer deep learning |
| url | https://www.frontiersin.org/articles/10.3389/fbioe.2024.1504249/full |
| work_keys_str_mv | AT yizhipan amutualinclusionmechanismforpreciseboundarysegmentationinmedicalimages AT junyixin amutualinclusionmechanismforpreciseboundarysegmentationinmedicalimages AT junyixin amutualinclusionmechanismforpreciseboundarysegmentationinmedicalimages AT tianhuayang amutualinclusionmechanismforpreciseboundarysegmentationinmedicalimages AT siqili amutualinclusionmechanismforpreciseboundarysegmentationinmedicalimages AT leminhnguyen amutualinclusionmechanismforpreciseboundarysegmentationinmedicalimages AT teeradajracharak amutualinclusionmechanismforpreciseboundarysegmentationinmedicalimages AT kaili amutualinclusionmechanismforpreciseboundarysegmentationinmedicalimages AT kaili amutualinclusionmechanismforpreciseboundarysegmentationinmedicalimages AT guanqunsun amutualinclusionmechanismforpreciseboundarysegmentationinmedicalimages AT guanqunsun amutualinclusionmechanismforpreciseboundarysegmentationinmedicalimages AT guanqunsun amutualinclusionmechanismforpreciseboundarysegmentationinmedicalimages AT yizhipan mutualinclusionmechanismforpreciseboundarysegmentationinmedicalimages AT junyixin mutualinclusionmechanismforpreciseboundarysegmentationinmedicalimages AT junyixin mutualinclusionmechanismforpreciseboundarysegmentationinmedicalimages AT tianhuayang mutualinclusionmechanismforpreciseboundarysegmentationinmedicalimages AT siqili mutualinclusionmechanismforpreciseboundarysegmentationinmedicalimages AT leminhnguyen mutualinclusionmechanismforpreciseboundarysegmentationinmedicalimages AT teeradajracharak mutualinclusionmechanismforpreciseboundarysegmentationinmedicalimages AT kaili mutualinclusionmechanismforpreciseboundarysegmentationinmedicalimages AT kaili mutualinclusionmechanismforpreciseboundarysegmentationinmedicalimages AT guanqunsun mutualinclusionmechanismforpreciseboundarysegmentationinmedicalimages AT guanqunsun mutualinclusionmechanismforpreciseboundarysegmentationinmedicalimages AT guanqunsun mutualinclusionmechanismforpreciseboundarysegmentationinmedicalimages |