README.md
These scripts and data provide example code for the method described in:
Haskell et al. 2019, "Network Accelerated Motion Estimation and Reduction (NAMER): Convolutional neural network guided retrospective motion correction using a separable motion model"
To start run namer_recon.m in MATLAB. This script requires that keras with TensorFlow backend installed on your machine, as MATLAB will call a python script to evaluate a CNN built using keras.
This example was most recently tested using matlab 2017b, and uses that versions's syntax for optimization settings. etc.
Key scripts
namer_recon.m - This script performs the separable cost function version of the NAMER method (Eqn 3 in Haskell et al. 2019), and corresponds to the result shown in the bottom left of Figure 4-B in the paper.
run_namer_cnn.py - This script evaluates all of the patches for a given input image and returns the output of the motion artifact detecting CNN.
train_namer_cnn.py- This script is provided as an example of how the CNN was constructed and trained. The training data is not provided to run this script to save on space, but can be shared by emailing melissa.w.haskell@gmail.com.