Ultra-low Dose PET Imaging Challenge

This is an example of the PET imaging is used to recovere high-quality imaging from low dose scans imaging.

Prerequisities

The following dependencies are needed:

• numpy >= 1.11.1
• SimpleITK >=1.0.1
• pytorch-gpu ==1.10.0
• pandas >=0.20.1
• scikit-learn >= 0.17.1

How to Use

• 1、when download the all project,check out the data folder all csv,put your train data into same folder.or you can run lowdosePet2022data3dprepare.py to generate train data and validation data.
• 2、run lowdosePet_train.py for Unet3d training:make sure train data have effective path
• 3、run lowdosePet_inference.py for Unet3d inference:make sure validation data have effective path
• 4、run lowdosePet_inferencetest.py for Unet3d inference test:make sure test data have effective path

Result

1-2 dose predict result

1-4 dose predict result

1-10 dose predict result

1-20 dose predict result

1-50 dose predict result

1-100 dose predict result

• you can find the trained model in log folder.
• more detail and trained model can follow my WeChat Public article.

Contact

• https://github.com/junqiangchen
• email: 1207173174@qq.com
• Contact: junqiangChen
• WeChat Number: 1207173174
• WeChat Public number: 最新医学影像技术