Fig 1.
Sample image of brain tumor.
Fig 2.
Proposed automated system for brain tumor analysis.
Fig 3.
Overall framework for the proposed system.
Fig 4.
Basic CNN structure.
Fig 5.
Basic U-Net structure.
Fig 6.
Basic U-Net++ structure.
Fig 7.
Web application architecture.
Table 1.
Datasets.
Table 2.
CNN model structure for the tumor detection model with input images.
Table 3.
CNN model structure for the tumor detection model with input features.
Table 4.
2D U-Net model structure for tumor segmentation.
Table 5.
2D U-Net++ model structure for tumor segmentation.
Table 6.
3D U-Net model structure for tumor segmentation.
Table 7.
3D U-Net++ model structure for tumor segmentation.
Fig 8.
Home page.
Fig 9.
Login page.
Fig 10.
Registration page.
Fig 11.
PACS page.
Fig 12.
Tumor detection page.
Fig 13.
Tumor detection evaluation results page.
Fig 14.
Tumor detection show result page.
Fig 15.
Tumor detection evaluation feedback page.
Fig 16.
2D segmentation page.
Fig 17.
Evaluation results page.
Fig 18.
Show results page.
Fig 19.
3D segmentation page.
Fig 20.
3D segmentation results page.
Fig 21.
Feedback page.
Fig 22.
My evaluations page.
Table 8.
Tumor/Non-tumor detection with image and image features.
Fig 23.
Performance comparison for tumor/non-tumor detection task.
Fig 24.
Performance comparison for tumor segmentation task.
Fig 25.
Performance comparison for 2D and 3D tumor segmentation (average prediction Dice scores).
Table 9.
2D and 3D tumor segmentation.
Table 10.
Comparison of whole tumor segmentation with BRATS 2021.
Fig 26.
Sample segmentation outputs for 2D U-Net segmentation.
Fig 27.
Sample segmentation outputs for 2D U-Net++ segmentation.
Fig 28.
Sample segmentation outputs for 3D U-Net segmentation.
Fig 29.
Sample segmentation outputs for 3D U-Net++ segmentation.