NEURAL DYSFUNCTIONS AS A RESULT OF MILD TRAUMATIC BRAIN INJURY: EEG STUDY
Open Access
- Author:
- Cao, Cheng
- Graduate Program:
- Kinesiology
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- June 08, 2009
- Committee Members:
- Semyon Slobounov, Dissertation Advisor/Co-Advisor
Semyon Slobounov, Committee Chair/Co-Chair
Karl Maxim Newell, Committee Member
Peter Cm Molenaar, Committee Member
Richard Laurence Tutwiler, Committee Member - Keywords:
- EEG
Traumatic Brain Injury - Abstract:
- Mild traumatic brain injury (MTBI), commonly known as “concussion” is still one of the most confusing neurological disorders and the least understood injuries in athletics. The neuropsychological deficits after MTBI are commonly resolved within one week after injury but residual neurological impairments may persist. The EEG measures have been traditionally used to detect residual neural dysfunctions after MTBI. However, the conventional EEG measures have several shortcomings, yielding controversial and inconsistent results. It is feasible to hypothesize that advanced EEG research protocols can overcome these shortcomings and provide more clear information regarding the long lasting neural impairment in subjects suffering from MTBI. In this dissertation several novel EEG algorithms have been proposed and validated to verify that residual EEG abnormalities may be present in concussed individuals in absence of any signs and subjective symptoms of concussion. Specifically, our current findings have clearly documented that EEG entropy measures have significantly decreased in several brain regions as a result of MTBI. Moreover, the non-stationarity of EEG signals at occipital and parietal regions was found to be significantly reduced in MTBI subjects, overall suggesting the lack of adaptive strategies as a result of neurological dysfunction. In addition, using the ICA combined with sLORETA and graph theory methods we have found the alteration of long-distance connectivity of EEG signals in concussed individuals. We tried to predict the recovery rate from concussion based on initial baseline testing and the brain’s initial response to first concussion. Indeed, differential response to first concussion may be a predictor of the rate of recovery. In addition, we applied the classifier support vector machine (SVM) to achieve the automatic classification of MTBI. We have demonstrated that by using novel nonlinear features and advanced robust classifiers, the residual deficit of MTBI can be identified in absence of any subjective symptoms of concussion. We believe that this research contributes to our current understanding of concussion, in general, and the neural dysfunctions in MTBI patients that can be overlooked if conventional assessment tools are implemented