Machine Learning Based MTBI Classification Using Diffusion MRI

1) MTBI Classification
Mild Traumatic Brain Injury (MTBI) is a growing public health problem with an underestimated incidence of over one million people annually in the U.S. Neuropsychological tests are used to both assess the patient’s condition and to monitor patient progress. The first objective of this study is to use features extracted from MRI images taken shortly after injury to predict whether a person has MTBI or not.
2) Repeated Head Impacts Classification
Besides, even without frank concussion, long-term contact and competitive sport playing could lead to microscopic alteration, which is related to repeated head impacts (RHI). Diffusion metrics and machine learning methods are designed to validate the existence of such subtle change. With great interpretability of the designed model, the work is to shed light on the unclear relationship between diffusion metric/brain region and underlying pathology of RHI. 
3) Cognitive Score Prediction
Based on the diffusion metric, the third goal is to predict subjects’ cognitive performance in several different aspects (working memory, processing speed, cognitive flexibility, e.t.c).  

Within this three tasks, we need to determine the most effective features and corresponding prediction method. The main challenge is that we have only limited training data, from which we need to develop the prediction method that can be expected to provide accurate prediction results for unseen data. Self-supervised learning is current being investigate to solve such challenging. 

Sponsor:
This work is supported by the US Department of Defense and  National Institute of Health

Current Participants:
Junbo Chen, PhD candidate
Prof. Yao Wang
Dr. Yvonne Lui
Dr. Sohae Chung

Previous Participants:
Shervin Minaee, alumnus (PhD graduated 2019)
Hugh Wang, Researcher

Related Publications:

• Sohae Chung, Junbo Chen, Tianhao Li, Yao Wang, Yvonne Lui, “Investigating Brain White Matter in Football Players with and without Concussion Using a Biophysical Model from Multi-shell Diffusion MRI”, Neurology, 2021 (submitted).
• Sohae Chung, Junbo Chen, Tianhao Li, Yao Wang, Yvonne Lui. “White Matter Microstructural Alterations In Contact-Sport Athletes With And Without Concussion: A Multi-Shell Diffusion Study“, International Society for Magnetic Resonance in Medicine (ISMRM), 2021.
• Junbo Chen, Sohae Chung, Joseph Rath, Els Fieremans, YaoWang, Yvonne Lui. “Predicting Visual-Motor Functioning in Patients with Mild Traumatic Brain Injury from Multi-Shell Diffusion MRI using Gradient Boosting Tree”, Annual Meeting of Radiological Society of North America (RSNA), 2020.
• Xu, Tongda, et al. “Identification of relevant diffusion MRI metrics impacting cognitive functions using a novel feature selection method.” 2019 IEEE Signal Processing in Medicine and Biology Symposium (SPMB). IEEE, 2019.
• Minaee, Shervin, et al. “MTBI identification from diffusion MR images using bag of adversarial visual features.” IEEE transactions on medical imaging 38.11 (2019): 2545-2555.
• Shervin Minaee, Yao Wang, Anna Choromanska, Sohae Chung, Xiuyuan Wang, Els Fieremans, Steven Flanagan, Joseph Rath, Yvonne W Lui, “A Deep Unsupervised Learning Approach Toward MTBI Identification Using Diffusion MRI”, International Engineering in Medicine and Biology Conference, IEEE, 2018.
• Shervin Minaee, Siyun Wang, Yao Wang, Sohae Chung, Xiuyuan Wang, Els Fieremans, Steven Flanagan, Joseph Rath, Yvonne W Lui, “Identifying Mild Traumatic Brain Injury Patients From MR Images Using Bag of Visual Words,”, IEEE Signal Processing in Medicine and Biology Symposium, Dec 2017.
• Shervin Minaee, Yao Wang, Sohae Chung, Xiuyuan Wang, Els Fieremans, Steven Flanagan, Joseph Rath, Yvonne W Lui, “A Machine Learning Approach For Identifying Patients with Mild Traumatic Brain Injury Using Diffusion MRI Modeling”, ASFNR 11th Annual Meeting, Oct 2017.
• S Minaee, Y Wang, YW Lui, “Prediction of longterm outcome of neuropsychological tests of MTBI patients using imaging features,” Signal Processing in Medicine and Biology Symposium (SPMB), IEEE, Dec 2013.