Related Experiment Video
Updated: Jul 27, 2025
Quantitation of Protein Expression and Co-localization Using Multiplexed Immuno-histochemical Staining and Multispectral Imaging
Published on: April 8, 2016
MAS-UNet: a U-shaped network for prostate segmentation
YuQi Hong1, Zhao Qiu1, Huajing Chen2
1School of Computer Science and Technology, Hainan University, Haikou, China.
Abstract:
Prostate cancer is a common disease that seriously endangers the health of middle-aged and elderly men. MRI images are the gold standard for assessing the health status of the prostate region. Segmentation of the prostate region is of great significance for the diagnosis of prostate cancer. In the past, some methods have been used to segment the prostate region, but segmentation accuracy still has room for improvement. This study has proposed a new image segmentation model based on Attention UNet. The model improves Attention UNet by using GN instead of BN, adding dropout to prevent overfitting, introducing the ASPP module, adding channel attention to the attention gate module, and using different channels to output segmentation results of different prostate regions. Finally, we conducted comparative experiments using five existing UNet-based models, and used the dice coefficient as the metric to evaluate the segmentation result. The proposed model achieves dice scores of 0.807 and 0.907 in the transition region and the peripheral region, respectively. The experimental results show that the proposed model is better than other UNet-based models.
Insights
This study introduces an improved Attention UNet model for precise prostate MRI segmentation, crucial for early prostate cancer detection. The enhanced model significantly boosts segmentation accuracy in key prostate regions.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Oncology
Background:
- Prostate cancer poses a significant health risk to middle-aged and elderly men.
- Accurate segmentation of prostate magnetic resonance imaging (MRI) is vital for cancer diagnosis.
- Existing segmentation methods require further accuracy improvements.
Purpose of the Study:
- To develop a novel, highly accurate prostate MRI segmentation model.
- To enhance the Attention UNet architecture for superior performance.
Main Methods:
- Proposed a modified Attention UNet incorporating Group Normalization (GN), dropout, and an Atrous Spatial Pyramid Pooling (ASPP) module.
- Integrated channel attention into the attention gate module.
- Utilized distinct output channels for segmenting different prostate regions.
Main Results:
- The enhanced model achieved Dice scores of 0.807 for the transition zone and 0.907 for the peripheral zone.
- Comparative experiments demonstrated superior performance over five existing UNet-based models.
Conclusions:
- The proposed Attention UNet-based model offers improved accuracy for prostate MRI segmentation.
- This advancement holds promise for more reliable prostate cancer diagnosis and assessment.
