Enhancing Medical Image Classification With Context Modulated Attention and Multi-Scale Feature Fusion
This research proposes a multi-stage feature fusion network (MSFF) for medical image classification. In view of the problems existing in medical images, such as noise, diversity, and similarity among different classes, MSFF enhances the global context perception in the window partitioning framework...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10848071/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832583999025840128 |
|---|---|
| author | Renhan Zhang Xuegang Luo Junrui Lv Junyang Cao Yangping Zhu Juan Wang Bochuan Zheng |
| author_facet | Renhan Zhang Xuegang Luo Junrui Lv Junyang Cao Yangping Zhu Juan Wang Bochuan Zheng |
| author_sort | Renhan Zhang |
| collection | DOAJ |
| description | This research proposes a multi-stage feature fusion network (MSFF) for medical image classification. In view of the problems existing in medical images, such as noise, diversity, and similarity among different classes, MSFF enhances the global context perception in the window partitioning framework through Context Modulation Attention (CMA). Meanwhile, it extracts fine-grained local information via the multi-stage Contextual Information Refinement (CIR) module and gradually fuses multi-stage local and global features to generate richer semantic representations. The experimental results demonstrate that MSFF significantly outperforms existing methods in multiple performance metrics (including accuracy, precision, recall, F1-score, Matthews Correlation Coefficient (MCC), Kappa coefficient, Area Under the Curve (AUC), balanced accuracy, and geometric mean) on four datasets (Endoscopic Bladder Tissue, Kvasir, SARS-COV-2 Ct-Scan, and Thyroid Nodule), showing its excellent performance in the task of medical image classification. |
| format | Article |
| id | doaj-art-bbe3289c856f409f80e1ae37657b70e4 |
| institution | Kabale University |
| issn | 2169-3536 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj-art-bbe3289c856f409f80e1ae37657b70e42025-01-28T00:01:47ZengIEEEIEEE Access2169-35362025-01-0113152261524310.1109/ACCESS.2025.353235410848071Enhancing Medical Image Classification With Context Modulated Attention and Multi-Scale Feature FusionRenhan Zhang0Xuegang Luo1https://orcid.org/0000-0001-8240-5199Junrui Lv2https://orcid.org/0000-0002-6796-3819Junyang Cao3Yangping Zhu4Juan Wang5https://orcid.org/0009-0006-0852-5022Bochuan Zheng6School of Computer Science, China West Normal University, Nanchong, Sichuan, ChinaSchool of Mathematics and Computer Science, Panzhihua University, Panzhihua, Sichuan, ChinaSchool of Mathematics and Computer Science, Panzhihua University, Panzhihua, Sichuan, ChinaSchool of Computer Science, China West Normal University, Nanchong, Sichuan, ChinaDepartment of Radiology, Nanjiang TCM Hospital, Bazhong, Sichuan, ChinaSchool of Computer Science, China West Normal University, Nanchong, Sichuan, ChinaSchool of Computer Science, China West Normal University, Nanchong, Sichuan, ChinaThis research proposes a multi-stage feature fusion network (MSFF) for medical image classification. In view of the problems existing in medical images, such as noise, diversity, and similarity among different classes, MSFF enhances the global context perception in the window partitioning framework through Context Modulation Attention (CMA). Meanwhile, it extracts fine-grained local information via the multi-stage Contextual Information Refinement (CIR) module and gradually fuses multi-stage local and global features to generate richer semantic representations. The experimental results demonstrate that MSFF significantly outperforms existing methods in multiple performance metrics (including accuracy, precision, recall, F1-score, Matthews Correlation Coefficient (MCC), Kappa coefficient, Area Under the Curve (AUC), balanced accuracy, and geometric mean) on four datasets (Endoscopic Bladder Tissue, Kvasir, SARS-COV-2 Ct-Scan, and Thyroid Nodule), showing its excellent performance in the task of medical image classification.https://ieeexplore.ieee.org/document/10848071/Medical imagesglobal semanticslocal featurestransformercontext modulated attentionmulti-stage feature fusion network |
| spellingShingle | Renhan Zhang Xuegang Luo Junrui Lv Junyang Cao Yangping Zhu Juan Wang Bochuan Zheng Enhancing Medical Image Classification With Context Modulated Attention and Multi-Scale Feature Fusion IEEE Access Medical images global semantics local features transformer context modulated attention multi-stage feature fusion network |
| title | Enhancing Medical Image Classification With Context Modulated Attention and Multi-Scale Feature Fusion |
| title_full | Enhancing Medical Image Classification With Context Modulated Attention and Multi-Scale Feature Fusion |
| title_fullStr | Enhancing Medical Image Classification With Context Modulated Attention and Multi-Scale Feature Fusion |
| title_full_unstemmed | Enhancing Medical Image Classification With Context Modulated Attention and Multi-Scale Feature Fusion |
| title_short | Enhancing Medical Image Classification With Context Modulated Attention and Multi-Scale Feature Fusion |
| title_sort | enhancing medical image classification with context modulated attention and multi scale feature fusion |
| topic | Medical images global semantics local features transformer context modulated attention multi-stage feature fusion network |
| url | https://ieeexplore.ieee.org/document/10848071/ |
| work_keys_str_mv | AT renhanzhang enhancingmedicalimageclassificationwithcontextmodulatedattentionandmultiscalefeaturefusion AT xuegangluo enhancingmedicalimageclassificationwithcontextmodulatedattentionandmultiscalefeaturefusion AT junruilv enhancingmedicalimageclassificationwithcontextmodulatedattentionandmultiscalefeaturefusion AT junyangcao enhancingmedicalimageclassificationwithcontextmodulatedattentionandmultiscalefeaturefusion AT yangpingzhu enhancingmedicalimageclassificationwithcontextmodulatedattentionandmultiscalefeaturefusion AT juanwang enhancingmedicalimageclassificationwithcontextmodulatedattentionandmultiscalefeaturefusion AT bochuanzheng enhancingmedicalimageclassificationwithcontextmodulatedattentionandmultiscalefeaturefusion |