MRI Reconstruction

A project developed as a part of Praktikum Machine Learning in Medical Imaging at the Technical University of Munich by

• Devendra Vyas
• Fabian Groeger
• Maik Dannecker
• Viktor Studenyak

Mentored by : Shahrooz Roohi

This work was used in the Multi-channel MR Image Reconstruction Challenge (MC-MRRec) as part of the 2020 Medical Imaging with Deep Learning (MIDL) conference.

How to Run

To run on your system simply go to the root directory and use the run scrip as:

my_user@my_device: ./run.sh --model=WNET

Or you can directly run the main_train.py using:

my_user@my_device: python main_train.py --model=WNET

The argument --model can take either of the 2 values {'WNET', 'RL'}