Using U-Net models in deep learning for brain tumor detection from MRI scans
Tumor diseases in the nervous system are both dangerous and complex. Magnetic Resonance Imaging (MRI) is crucial for detecting brain disease; however, identifying the presence of tumors from these is time-consuming and requires a professional doctor. Utilizing deep learning for tumor detection in M...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Can Tho University Publisher
2024-10-01
|
| Series: | CTU Journal of Innovation and Sustainable Development |
| Subjects: | |
| Online Access: | https://ctujs.ctu.edu.vn/index.php/ctujs/article/view/1165 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849688883547078656 |
|---|---|
| author | Minh Khiem Nguyen Phuoc Huy Tran Tan Tai Phan |
| author_facet | Minh Khiem Nguyen Phuoc Huy Tran Tan Tai Phan |
| author_sort | Minh Khiem Nguyen |
| collection | DOAJ |
| description |
Tumor diseases in the nervous system are both dangerous and complex. Magnetic Resonance Imaging (MRI) is crucial for detecting brain disease; however, identifying the presence of tumors from these is time-consuming and requires a professional doctor. Utilizing deep learning for tumor detection in MRI images can reduce waiting times and enhance detection accuracy. We propose a method employing two U-Net models: ResNeXt-50 and EfficientNet architectures, integrated with a Feature Pyramid Network (FPN) for segmenting brain tumor. The models were trained on the BraTS 2021 dataset, consisting of 3,929 MRI scan images with 3,929 corresponding masks, divided into training, testing, and evaluation sets in a 70:15:15 ratio. The results indicate that the hybrid model, which combines EfficientNet and FPN, delivers superior performance, with an average Intersection over Union (IoU) accuracy of 0.90 on the test set compared to 0.50 for ResNeXt-50, and Dice accuracy of 0.92 compared to 0.66 for ResNeXt-50. Furthermore, we developed a web application that implements the EfficientNet with FPN model, facilitating convenient tumor detection from uploaded MRI images for doctors.
|
| format | Article |
| id | doaj-art-a6c0a6cb9cb743faa579b350098bad21 |
| institution | DOAJ |
| issn | 2588-1418 2815-6412 |
| language | English |
| publishDate | 2024-10-01 |
| publisher | Can Tho University Publisher |
| record_format | Article |
| series | CTU Journal of Innovation and Sustainable Development |
| spelling | doaj-art-a6c0a6cb9cb743faa579b350098bad212025-08-20T03:21:50ZengCan Tho University PublisherCTU Journal of Innovation and Sustainable Development2588-14182815-64122024-10-0116Special issue: ISDS10.22144/ctujoisd.2024.326Using U-Net models in deep learning for brain tumor detection from MRI scansMinh Khiem Nguyen0Phuoc Huy Tran1Tan Tai Phan2College of Information and Communication Technology, CanTho University, CanTho, Vietnam VNPT Soc Trang, Soc Trang, Vietnam College of Information and Communication Technology, CanTho University, CanTho, Vietnam Tumor diseases in the nervous system are both dangerous and complex. Magnetic Resonance Imaging (MRI) is crucial for detecting brain disease; however, identifying the presence of tumors from these is time-consuming and requires a professional doctor. Utilizing deep learning for tumor detection in MRI images can reduce waiting times and enhance detection accuracy. We propose a method employing two U-Net models: ResNeXt-50 and EfficientNet architectures, integrated with a Feature Pyramid Network (FPN) for segmenting brain tumor. The models were trained on the BraTS 2021 dataset, consisting of 3,929 MRI scan images with 3,929 corresponding masks, divided into training, testing, and evaluation sets in a 70:15:15 ratio. The results indicate that the hybrid model, which combines EfficientNet and FPN, delivers superior performance, with an average Intersection over Union (IoU) accuracy of 0.90 on the test set compared to 0.50 for ResNeXt-50, and Dice accuracy of 0.92 compared to 0.66 for ResNeXt-50. Furthermore, we developed a web application that implements the EfficientNet with FPN model, facilitating convenient tumor detection from uploaded MRI images for doctors. https://ctujs.ctu.edu.vn/index.php/ctujs/article/view/1165EfficiencyNet FPN, ResneXt-50, Tumor disease |
| spellingShingle | Minh Khiem Nguyen Phuoc Huy Tran Tan Tai Phan Using U-Net models in deep learning for brain tumor detection from MRI scans CTU Journal of Innovation and Sustainable Development EfficiencyNet FPN, ResneXt-50, Tumor disease |
| title | Using U-Net models in deep learning for brain tumor detection from MRI scans |
| title_full | Using U-Net models in deep learning for brain tumor detection from MRI scans |
| title_fullStr | Using U-Net models in deep learning for brain tumor detection from MRI scans |
| title_full_unstemmed | Using U-Net models in deep learning for brain tumor detection from MRI scans |
| title_short | Using U-Net models in deep learning for brain tumor detection from MRI scans |
| title_sort | using u net models in deep learning for brain tumor detection from mri scans |
| topic | EfficiencyNet FPN, ResneXt-50, Tumor disease |
| url | https://ctujs.ctu.edu.vn/index.php/ctujs/article/view/1165 |
| work_keys_str_mv | AT minhkhiemnguyen usingunetmodelsindeeplearningforbraintumordetectionfrommriscans AT phuochuytran usingunetmodelsindeeplearningforbraintumordetectionfrommriscans AT tantaiphan usingunetmodelsindeeplearningforbraintumordetectionfrommriscans |