Validating Novel Brain Imaging Biomarkers for Classifying Mild Traumatic Brain Injury and Subsequent Risks of Alzheimer s Disease in Gulf War Veterans
Abstract
Traumatic brain injury (TBI) is defined as trauma or acceleration of the head and brain that results in alteration or loss of consciousness (LOC). The severity of TBI is classified into mild, moderate, or severe based on the presence and duration of LOC, alteration of consciousness, and post-traumatic amnesia. Neuroimaging of military personnel with a history of TBI event has provided important clues about the mechanisms of injury and possible relationship with AD conversion later in life. Preliminary results from the longitudinal Department of Defense Alzheimer’s Disease Neuroimaging Initiative (DOD ADNI) study have shown that cerebrospinal fluid and neuroimaging measures collected from Veterans with TBI resembles AD pathology. Among different stages of TBI, mTBI is the most common type of TBI affecting military personnel. The Defense and Veterans Brain Injury Center estimated that 10%-20% of those returning home after combat exposure may have sustained mTBI, and the rate is even higher (~30%) in the large, longitudinally followed Ft. Devens cohort and in the Boston Gulf War Illness Consortium cohort of Gulf War (GW) Veterans. Recent survey data found that even mTBI without LOC was associated with more than a two-fold increase in the risk of dementia in Veteran populations. This may suggest that mTBI might have long-term neurodegenerative consequences. However, detecting and evaluating ongoing pathology in mTBI has been a challenging issue due to lack of standardized imaging and analytical methods and the right population with a history of mTBI and the age range to begin to see AD development. This also limits our understanding of the underlying neuropathobiological progression between mTBI and AD. Magnetic resonance imaging (MRI) has been successfully applied to identify neurological changes prior to the AD onset. Risk of AD has recently been predicted by various MRI markers, including morphological, connectional, and abnormal signal patterns in the brain. Dr. Koo, the Principal Investigator (PI) of the proposed project, has been working on applying novel MRI measures to study illness symptoms in GW Veterans. Recently, the PI and his research team found novel MRI markers associated with mTBI. Those markers also had associations with chronic symptoms of pain, fatigue, and sleep quality. The central objective in this project is to investigate the utility of the novel MRI markers in predicting progressive neurological damage in Veterans with mTBI, estimate the probability of AD progression based on the neuroimaging proximity measures between mTBI and AD prognostic imaging markers, and build a computational model to provide accurate classification of mTBI and prediction of subsequent risk of AD. To bring novel insights and technologies for Veteran populations, this project is designed to incorporate the PI s recent findings with up-to-date neuroimaging biomarkers for AD prognosis. We will utilize the abundant amount of biomarker data, which has already been collected from the large, multi-site CDMRP-funded Boston Gulf War Illness Consortium (GWIC), and the follow-up data, which will be collected from a recently funded GW longitudinal MRI project with the same Veterans (Dr. Sullivan is the PI of these GW studies). Recently, MRI has been successfully applied for to identify neurological changes 3~8 years prior to the onset of AD. From the ADNI database, baseline MRI scans on more than 200 subjects who converted to AD in their later time point observations will be combined with traditional prodromal AD classifications for building reference information for machine learning analyses. These two different ground-breaking cohort study datasets (ADNI, GWIC) will be combined in computer algorithms (jointly embedded) to provide time and cost-efficient data to answer the question of who is likely to develop AD after mTBI. We will utilize a machine learning framework, which is derived from artificial intelligence te
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Mar 10, 2021
- Source ID
- W81XWH2010236
Entities
People
- Bang-Bon Koo
Organizations
- Boston University Medical Campus
- United States Army