What is it about?
The proportion of pancreatic and its tumors in the abdominal cavity is extremely small, making it difficult for deep learning techniques to accurately segment them. We have developed a new 3D neural network model based on 3D U-Net to segment pancreatic tumors in CT images. The experimental results demonstrate the effectiveness of our model.
Featured Image
Photo by Kevin Ku on Unsplash
Why is it important?
The 3D neural network model we designed directly segments pancreatic tumors in 3D CT images. While improving segmentation accuracy, it also avoids the loss of inter layer information in the 2D model segmentation process.
Perspectives
I am writing this article because I am interested in the field of combining deep learning techniques with medical imaging. I hope readers who have read this article can have a better understanding of pancreatic tumors.
Jiakun Deng
Wuhan Polytechnic University
Read the Original
This page is a summary of: Pancreatic Tumor Segmentation Based on 3D U-Net with Densely Connected Atrous Spatial Pyramid Module and Attention Module, October 2023, ACM (Association for Computing Machinery),
DOI: 10.1145/3644116.3644127.
You can read the full text:
Contributors
The following have contributed to this page
