Date of Award


Degree Type



Chemical and Biomedical Engineering

First Advisor

Yue, Guang H.

Subject Headings

Amyotrophic lateral sclerosis, Fractals, Brain -- Degeneration, Brain -- Magnetic resonance imaging, Amyotrophic lateral sclerosis (ALS), degeneration, white matter (WM), fractal dimension (FD), skeleton, surface, general structure, magnetic resonance imaging (MRI)


Amyotrophic lateral sclerosis (ALS) is recognized as a motor neuron disorder affecting the structure and function of the brain and neuromuscular system. Very little is known, however, that the nervous system degeneration is dependent on disease phenotypes of ALS. The purpose of this study was to determine the degree of brain white matter (WM) structure degeneration in four ALS patient groups characterized by their clinical signs and neuroimaging measurements. Fractal dimension (FD) of three-dimensional (3D) brain WM images was quantitatively analyzed to evaluate the WM structural complexity, including complexity levels of the WM skeleton, surface and general structures in ALS patients and control subjects. A total of 100 participants were assigned into five groups: ALS patients with frontotemporal dementia (ALS-FTD, n=20), ALS patients with predominantly upper motor neuron (UMN) signs and hyperintensity MRI signals on corticospinal tract (CST) (UMN-CST+, n=20), ALS patients with predominantly UMN signs but without hyperintensity signal on CST (UMN-CST-, n=27), ALS patients with an equal amount of UMN and lower motor neuron (LMN) signs (ALS-classic, n=22), and a neurological control group (n=11). The brain was extracted from head images by using the FSL package, and the WM was segmented from and the brain images before being separate into WM images of the left and right hemispheres. Subsequently, skeletons of the WM bundles were obtained using a 3-dimensional thinning method. FD analysis was applied onto three forms of the WM structure: skeleton, surface, and general structure. FD of the skeletons and general structure in ALS-FTD patients was significantly smaller (P<0.05 - P>0.01) than the controls and UMN-CST+, and ALS-classic patients. The FD of UMN-CST+ patients was significantly larger (P<0.05 - P>0.01) than UMN-CST- and ALS-classic patients. These results suggest that the complexity level of brain WM network is dependent on ALS disease phenotypes and ALS patients with dementia suffer the worst brain WM stru