Your Donations in Action
Quantifying White Matter Degeneration to Identify Imaging Biomarkers for Dementia
There is considerable overlap in imaging features between aging white matter and dementia such as Alzheimer’s disease (AD). Identifying a robust, quantitative marker of white matter complexity has the potential to provide a metric to assess the age-appropriateness of white matter disorganization, enabling the early identification of neurodegenerative disorders.
In his 2017 RSNA Research Resident Grant study, “Quantifying White Matter Changes in Aging and Dementia Through Sparse Encoding of Diffusion-weighted MRI of the Brain,” Vishal Patel, MD, PhD, sought to quantify changes in normal aging white matter, mild cognitive impairment and AD using a recently validated imaging metric derived from applying a sparse encoding scheme to the raw diffusion-weighted signal.
Dr. Patel and colleagues performed a unique analysis of data from the multicenter Alzheimer’s Disease Neuroimaging Initiative. Each diffusion-weighted data set was encoded using the K-SVD algorithm and the sparsity of the results were quantified by the Gini coefficient. This metric was statistically compared across the clinical groups to determine age-appropriate values and to evaluate for significant deviations in the disease states. In addition, the Gini coefficient was evaluated against subject performance on neurocognitive tests to assess for the presence of a predictive relationship.
“Ultimately, we hope that by applying this technique, we may discover properties that enable the noninvasive detection of Alzheimer’s dementia at an early stage, when current and future treatment options are likely to have the best efficacy in reducing disease-associated morbidity and delaying the progression of clinical symptoms,” Dr. Patel said.