/MRI_Motion_Classification

Generate and detect motion artifact on MRI images using Python.

Primary LanguagePython

MRI_Motion_Classification

MRI_Motion_Classification

Generating synthetic motion on MRI images based on Cartesian k-space sampler: Utils/MotionUtils/GenerateMotion.py More k-space sampling trajectory can be designed just by interfacing CartesianSampler.py

The generated motion is slice-dependent. In other words, for each slice it rotate the rigid 3d volume and apply k-space sampling in appropriate time.

For training the model fix directory in DataGenerator.py

Reference

Mohebbian, M., Walia, E., Habibullah, M., Stapleton, S. and Wahid, K.A., 2021. Classifying MRI motion severity using a stacked ensemble approach. Magnetic Resonance Imaging, 75, pp.107-115.