DeepPET: An Intelligent Brain PET Device with Low Radiation and High Imaging Quality

Yuezhe Yang, Qinggaoyuan Zhou, Sitong Zhou, Sheng Yan, Wenjie Cai, Ruixi Feng, Jinyang Zhao, Nuo Chen, Boyu Yang, Xiangzhuo Zhao, Ran Wei, Kaiwen Zhu, Haoliang Duan, Dexin Yang, Yang He, Yaqian Wang, Huatao Yang, Liwen Wang, Ying Zhou, Xingguo Lv, Quan Zhang, Hui Zhang, Yonglin Chen, Dailei Zhang, Xingbo Dong†, Siyi Song†, Huabin Wang†, Zhe Jin†
†Project Supervisor
Anhui Provincial International Joint Research Center for Advanced Technology in Medical Imaging, Anhui University
Code

Abstract

This project improves PET image reconstruction by combining physical models with deep learning. Our goal is to enhance imaging accuracy, boost efficiency, and lower radiation exposure. We optimize cost, speed, and quality by using fewer sensors, refining equipment, and applying advanced reconstruction methods.

Our research focuses on efficient PET reconstruction using both model- and data-driven approaches. We also explore generative models to remove artifacts.

Equipment manufacturers and hospitals will benefit from these innovations. Despite competition, our advanced techniques will secure a strong market position. This work will drive progress in medical imaging and healthcare.

Highlights

  • Fuse physics models with deep learning for accurate efficient PET reconstruction.
  • Apply generative models to enhance low-count PET images and remove artifacts.
  • Optimize low-dose reconstruction for faster imaging and reduced radiation.

Video

Gallery

Honor

Patent

Awards

Citation

@misc{PET-Project,
              title = {DeepPET: An Intelligent Brain PET Device with Low Radiation and High Imaging Quality},
              author = {Yuezhe Yang},
              year = {2024},
              url = {https://github.com/Bean-Young/PET-Project},
}