Related Experiment Video
Updated: Apr 26, 2026
04:48
Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
Published on: July 5, 2024
993
Enhanced ResU-Net for brain tumor segmentation using EfficientNetB0, channel attention, and ASPP
Majid Behzadpour1, Ebrahim Azizi1, Bengie L Ortiz2
1Department of Electrical and Computer Engineering, Texas Tech University, Lubbock, TX 79409, United States of America.
Biomedical Physics & Engineering Express
|April 24, 2026
Summary
This study introduces an enhanced ResU-Net model for precise brain tumor segmentation, improving diagnostic accuracy. The novel architecture significantly outperforms existing methods on benchmark datasets.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Computational Biology
Background:
- Accurate brain tumor segmentation is vital for clinical decision-making.
- Existing segmentation methods face challenges with tumor variability.
Purpose of the Study:
- To develop an enhanced ResU-Net architecture for automatic brain tumor segmentation.
- To improve the accuracy and efficiency of brain tumor segmentation using deep learning.
Main Methods:
- Integration of EfficientNetB0 encoder for efficient feature extraction.
- Incorporation of channel attention mechanism to focus on relevant tumor features.
- Utilization of Atrous Spatial Pyramid Pooling (ASPP) for multiscale contextual learning.
Main Results:
- The proposed model achieved a Dice Similarity Coefficient (DSC) of 0.903 and a Hausdorff distance 95th percentile (HD95) of 9.43 for whole tumor segmentation on the BraTS 2020 dataset.
- The model demonstrated superior performance compared to baseline ResU-Net and its EfficientNet variant.
- Competitive results were obtained against state-of-the-art methods, especially for whole tumor and tumor core segmentation.
Conclusions:
- Combining EfficientNetB0, channel attention, and ASPP significantly enhances brain tumor segmentation.
- The developed model shows promise for clinical applications and other medical image segmentation tasks.
- This approach offers a robust solution for accurate and efficient brain tumor segmentation.