danielfrees / braingraphML

Project to investigate brain connectivity from rsfMRI images using graph theory. ML Classifier to classify TBI vs. Healthy Controls.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

TBI BrainGraph ML Project

Project banner (patient 1 spring layout brain graph)

Figure: Patient 1 Brain Graph: Spring Layout

GitHub last commit

Purpose

I am designing/leading my own primary research project (under the guidance of Dr. Kevin Bickart) to:
•Model mild traumatic brain injury patients brains as graphs
•Explore differences in topological properties (GT metrics) between traumatic brain injury patients and healthy controls
•Hypothesize important graph theory metrics to feed as features to an ML algorithm
•Derive an ML classifier to classify injured brains from healthy ones using graph theoretical metrics as features

The majority of my online-publishable work on the project currently resides in 'T1_Explore.ipynb'.

Note: image files, CONN extraction, and MATLAB extraction processes not uploaded here to protect privacy of HIPAA protected patient information. All data comes from real subjects.

If you have any questions about this project, would like to use/modify my code for a project you are working on, or would otherwise like to get in contact with me, my email is: danielfrees@g.ucla.edu

About

Project to investigate brain connectivity from rsfMRI images using graph theory. ML Classifier to classify TBI vs. Healthy Controls.

License:Apache License 2.0


Languages

Language:Jupyter Notebook 99.7%Language:Python 0.3%