This is an implementation of paper "S. Xue and R. Bai. 2D Probabilistic Undersampling Pattern Optimization for MR Image Reconstruction. IEEE Transactions on Medical Imaging, 2020, under review."
- OS: Ubuntu Linux kernel 4.15.0-76-generic x86_64
- CPU: Intel Xeon(R) Platinum 8163 CPU @ 2.50GHz x 96
- Memory: 503 GB
- GPU: NVIDIA Tesla V100 (32 GB) x 4
- Python 2.7.15
- Tensorflow 1.13.1
- Keras 2.2.4
- CUDA Toolkit 10.0.130
- CUDNN 7.6.5
- OpenCV 2.4.11
- Numpy 1.16.5
- 3T MR images
- MICCAI images
python main.py --gpu=0 --run=train --dataset=MRI_3T --model=VDSR
python main.py --gpu=0 --run=validate --dataset=MRI_3T --model=VDSR
Ph.D., Shengke Xue
College of Information Science and Electronic Engineering
Zhejiang University, Hangzhou, P.R. China