New research uncovers link between neurodegenerative disease and changes in brain shape
April 14, 2023
Researchers from the Biostatistics and Imaging Genomics analysis lab – Statistics & Signal (BIG-S2) at UNC-Chapel Hill have uncovered new information that demonstrates how neurodegenerative diseases, such as Alzheimer’s disease, can significantly accelerate changes in subcortical areas of the brain in those ages 60-75 in comparison to normal aging processes.
Through a series of brain scans, size changes were visible in the lateral ventricle, which contains and helps circulate cerebrospinal fluid, and the hippocampus, which supports limbic system processing of emotion, memory and behavior.
These findings were published recently as a discussion paper in the Journal of American Statistics Association (JASA), one of the most prestigious statistical journals which only publishes a few discussion papers each year. In the paper, the research team introduced a new method called longitudinal elastic shape analysis (LESA) for examining changes in specific brain regions over time.
This comprehensive framework consists of five key components: subcortical surface extraction, elastic shape analysis, principal components analysis (PCA) of shapes, continuous shape trajectory fitting, and shape-trajectory-on-scalar regression.
Dr. Hongtu Zhu
“The innovative LESA framework has enhanced our ability to visualize and comprehend spatiotemporal changes in complex brain structures, such as the hippocampus,” said Hongtu Zhu, PhD, professor of biostatistics at the UNC Gillings School of Global Public Health and co-author on the study. “This groundbreaking approach holds promise for shedding light on the effects of Alzheimer’s Disease and advancing the development of more effective treatments.”
The team applied this innovative framework to study brain scans from magnetic resonance imaging (MRI) in 2,275 individuals, analyzing a total of 9,628 shape surfaces. They found that the atrophy of subcortical regions begins early in life, around the age of 30, and accelerates after 60 years old. Furthermore, findings show that Alzheimer’s disease further speeds up this shrinkage in comparison to normal aging for those between 60 to 70 years old.
This figure depicts the left hippocampus shape change from age 60 to 90 in a hypothetical married female with 16 years education and one copy of APOE4 under three conditions, Alzheimer’s disease (AD), mild cognitive impairment (MCI), and healthy aging (NC).
The LESA framework enables researchers to accurately identify shape changes on the subcortical surfaces. They discovered that atrophy of the hippocampus associated with Alzheimer’s disease primarily occurs in the back of the organ, which is where crucial parts, known as subfields CA1, CA1, CA2 and CA4, are located. In addition, both Alzheimer’s disease and genetic risk factors — specifically the presence of two genetic alleles called ApoE4 — contribute to more severe atrophy of subcortical regions, such as hippocampus and lateral ventricle, during the aging process.
The study was a joint effort between Florida State University (FSU), the UNC College of Arts and Sciences and the Gillings School. The co-authors of the study include Zhengwu Zhang, PhD, assistant professor in the Department of Statistics and Operations Research at UNC; Yuexuan Wu, a doctoral student in the Department of Statistics at FSU; Di Xiong, a visiting doctoral student in biostatistics at UNC; Anuj Srivastava, PhD, a professor of statistics at FSU; and two biostatistics professors at the Gillings School, Joseph G. Ibrahim, PhD, and Hongtu Zhu, PhD. Professors Srivastava and Zhu served as the corresponding authors for this study.
Contact the UNC Gillings School of Global Public Health communications team at sphcomm@unc.edu.
