AUTHOR=Hong YuQi , Qiu Zhao , Chen Huajing , Zhu Bing , Lei Haodong 

TITLE=MAS-UNet: a U-shaped network for prostate segmentation

JOURNAL=Frontiers in Medicine

VOLUME=Volume 10 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/medicine/articles/10.3389/fmed.2023.1190659

DOI=10.3389/fmed.2023.1190659

ISSN=2296-858X

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 the segmentation accuracy still has room for improvement. This paper proposes a new image segmentation model based on Attention Unet. This model improves Attention Unet by using GN instead of BN, adding dropout to prevent overfitting, introducing aspp module, 3adding channel attention to the attention gate module, and using different channels to output segmentation results of different prostate regions. Finally, we conduct comparative experiments using five existing Unet-based models, and use the dice coefficient as the standard to evaluate the segmentation effect. 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.