Decoding brain tumor insights: Evaluating CAM variants with 3D U-Net for segmentation
Brain tumor segmentation is critical for effective diagnosis and treatment planning. While, conventional manual segmentation techniques are seen inefficient and variable, highlighting the need for automated methods. This study enhances medical image analysis, particularly in brain tumor segmentation...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Komunitas Ilmuwan dan Profesional Muslim Indonesia
2024-12-01
|
| Series: | Communications in Science and Technology |
| Subjects: | |
| Online Access: | https://cst.kipmi.or.id/journal/article/view/1477 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841560579190292480 |
|---|---|
| author | Dian Nova Kusuma Hardani Igi Ardiyanto Hanung Adi Nugroho |
| author_facet | Dian Nova Kusuma Hardani Igi Ardiyanto Hanung Adi Nugroho |
| author_sort | Dian Nova Kusuma Hardani |
| collection | DOAJ |
| description | Brain tumor segmentation is critical for effective diagnosis and treatment planning. While, conventional manual segmentation techniques are seen inefficient and variable, highlighting the need for automated methods. This study enhances medical image analysis, particularly in brain tumor segmentation by improving the explainability and accuracy of deep learning models, which are essential for clinical trust. Using the 3D U-Net architecture with the BraTS 2020 dataset, the study achieved precise localization and detailed segmentation with the mean recall values of 0.8939 for Whole Tumor (WT), 0.7941 for Enhancing Tumor (ET), and 0.7846 for Tumor Core (TC). The Dice coefficients were 0.9065 for WT, 0.8180 for TC, and 0.7715 for ET. By integrating explainable AI techniques, such as Class Activation Mapping (CAM) and its variants (Grad-CAM, Grad-CAM++, and Score-CAM), the study ensures high segmentation accuracy and transparency. Grad-CAM, in this case, provided the most reliable and detailed visual explanations, significantly enhancing model interpretability for clinical applications. This approach not only enhances the accuracy of brain tumor segmentation but also builds clinical trust by making model decisions more transparent and understandable. Finally, the combination of 3D U-Net and XAI techniques supports more effective diagnosis, treatment planning, and patient care in brain tumor management. |
| format | Article |
| id | doaj-art-b6421c4b079c48de8a0b96d558795e39 |
| institution | Kabale University |
| issn | 2502-9258 2502-9266 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | Komunitas Ilmuwan dan Profesional Muslim Indonesia |
| record_format | Article |
| series | Communications in Science and Technology |
| spelling | doaj-art-b6421c4b079c48de8a0b96d558795e392025-01-04T01:17:06ZengKomunitas Ilmuwan dan Profesional Muslim IndonesiaCommunications in Science and Technology2502-92582502-92662024-12-019226227310.21924/cst.9.2.2024.14771477Decoding brain tumor insights: Evaluating CAM variants with 3D U-Net for segmentationDian Nova Kusuma Hardani0Igi Ardiyanto1Hanung Adi Nugroho2Department of Electrical and Information Engineering, Universitas Gadjah Mada, Yogyakarta 55281, Indonesia; Department of Electrical Engineering, Universitas Muhammadiyah Purwokerto, Purwokerto 53182, IndonesiaDepartment of Electrical and Information Engineering, Universitas Gadjah Mada, Yogyakarta 55281, IndonesiaDepartment of Electrical and Information Engineering, Universitas Gadjah Mada, Yogyakarta 55281, IndonesiaBrain tumor segmentation is critical for effective diagnosis and treatment planning. While, conventional manual segmentation techniques are seen inefficient and variable, highlighting the need for automated methods. This study enhances medical image analysis, particularly in brain tumor segmentation by improving the explainability and accuracy of deep learning models, which are essential for clinical trust. Using the 3D U-Net architecture with the BraTS 2020 dataset, the study achieved precise localization and detailed segmentation with the mean recall values of 0.8939 for Whole Tumor (WT), 0.7941 for Enhancing Tumor (ET), and 0.7846 for Tumor Core (TC). The Dice coefficients were 0.9065 for WT, 0.8180 for TC, and 0.7715 for ET. By integrating explainable AI techniques, such as Class Activation Mapping (CAM) and its variants (Grad-CAM, Grad-CAM++, and Score-CAM), the study ensures high segmentation accuracy and transparency. Grad-CAM, in this case, provided the most reliable and detailed visual explanations, significantly enhancing model interpretability for clinical applications. This approach not only enhances the accuracy of brain tumor segmentation but also builds clinical trust by making model decisions more transparent and understandable. Finally, the combination of 3D U-Net and XAI techniques supports more effective diagnosis, treatment planning, and patient care in brain tumor management.https://cst.kipmi.or.id/journal/article/view/1477brain tumor segmentation3d u-netexplainable aiclass activation mappingdeep learningmedical imaging |
| spellingShingle | Dian Nova Kusuma Hardani Igi Ardiyanto Hanung Adi Nugroho Decoding brain tumor insights: Evaluating CAM variants with 3D U-Net for segmentation Communications in Science and Technology brain tumor segmentation 3d u-net explainable ai class activation mapping deep learning medical imaging |
| title | Decoding brain tumor insights: Evaluating CAM variants with 3D U-Net for segmentation |
| title_full | Decoding brain tumor insights: Evaluating CAM variants with 3D U-Net for segmentation |
| title_fullStr | Decoding brain tumor insights: Evaluating CAM variants with 3D U-Net for segmentation |
| title_full_unstemmed | Decoding brain tumor insights: Evaluating CAM variants with 3D U-Net for segmentation |
| title_short | Decoding brain tumor insights: Evaluating CAM variants with 3D U-Net for segmentation |
| title_sort | decoding brain tumor insights evaluating cam variants with 3d u net for segmentation |
| topic | brain tumor segmentation 3d u-net explainable ai class activation mapping deep learning medical imaging |
| url | https://cst.kipmi.or.id/journal/article/view/1477 |
| work_keys_str_mv | AT diannovakusumahardani decodingbraintumorinsightsevaluatingcamvariantswith3dunetforsegmentation AT igiardiyanto decodingbraintumorinsightsevaluatingcamvariantswith3dunetforsegmentation AT hanungadinugroho decodingbraintumorinsightsevaluatingcamvariantswith3dunetforsegmentation |