Source codes and trained networks described in the paper: Learning-based single-step quantitative susceptibility mapping reconstruction without brain extraction
Link: https://www.sciencedirect.com/science/article/pii/S1053811919306469
###Environmental Requirements:
Python 3.6
Tensorflow 1.15.0
Keras 2.2.5
###Files descriptions:
AutoQSM contains the following folders:
code: It contains the source codes for training and testing.
models/vnet/model_final_1.hdf5: Weights of AutoQSM.
test_data: one data for testing.
###Usage
##Test
You can run code/test.py to test the network performance on the provided data. The output will be in the 'results' directory
For test on your own data. You can use "data_predict" function as shown in test.py files
##train
If you want to train AutoQSM by yourself. code/train.py can be used as a reference.