MRIMath is a project dedicated towards developing a robust, automated brain tumor segmentation system. This repository contains all of my M.S. research, which can be a lot. Due to the nature of research, this is quite messy. However, I intend on cleaning this up significantly once we start making serious progress in one direction. In the meantime, if you have any questions or concerns about the project, feel free to email me at [email protected]. I hope that some of what has been written is useful to others!
- Python 3.x
- Tensorflow
- Keras
- Numpy
- OpenCV
- Nibabel
- Tensorlayer
- Nipype
- Matplotlib
Pull down the repository, and run the 'configure' script to install all python dependencies and download the BraTS 2018 data. From there, it should be a sandbox environment to explore various approaches to brain tumor segmentation and new data preprocessing techniques.
Rowan University's Deep Learning research lab, lead by Dr. Nidhal Bouaynaya and Dr. Ghulam Rasool, primarily specializes in applying deep learning to segmentation tasks. While the applications are typically related to brain MRI or potholes, we hope that our work can be used by other researchers and hobbyists.