BIO

My name is Mattan Pelah and I was born in Cambridge, England, and lived there until moving to South Florida at the age of 10. I studied at Florida Atlantic University High School, where I took, exclusively, university classes starting in my sophomore year of high school. I was always interested in the brain and began as a neuroscience major. However, after discovering the emergence of artificial intelligence in today's technological world, as well as the ever-increasing reliance on computational methods in science, I switched to studying computer science. I plan to further study the brain in graduate school and look forward to a career associated with investigating cognition, neural computation, and neurodegeneration. I am grateful to my UROP mentor, Dr. Hurdal, for working with me on this brain research project, and I look forward to presenting our findings at this symposium!

Mathematical and computational methods for comparing anatomical structures in MRI brain scans of Alzheimer's Disease and healthy controls

Abstract

According to the Centers for Disease Control (CDC), as many as 5.8 million Americans are afflicted with Alzheimer’s Disease (AD), a condition that results in significant and progressive cognitive impairments. While research on AD progresses steadily, the total number of cases is projected to nearly triple to 14 million by the year 2060 [1]. This project applies a mathematical method derived from planar geometry, called ‘circle packing.’ Anatomical structures from the human brain are extracted from Magnetic Resonance Imaging (MRI) data from AD patients and compared with healthy controls. Conformal (angle preserving) maps are then created using mathematical techniques. The resulting conformal maps can be used to define a coordinate system from which the features representing anatomical landmarks can be described and used to identify significant disparities between the two sample groups. Using this innovative method of studying brain anatomy, we hope to make significant new discoveries about characteristic anatomical features affected by AD. The methods being developed may potentially be applied to analyses of brains affected by other common neurodegenerative diseases, such as Parkinson’s disease, and in various brain regions (e.g., [2]).

[1] Alzheimer’s Disease and Healthy Aging: Alzheimer's Disease and Related Dementias. Centers for Disease Control (CDC). https://www.cdc.gov/aging/aginginfo/alzheimers.htm#:~:text=In%202020%2C%20as%20many%20as,were%20living%20with%20Alzheimer's%20disease.&text=Younger%20people%20may%20get%20Alzheimer's,14%20million%20people%20by%202060. Accessed on 2/22/2022.
[2] Hurdal, M.K., Bowers, P.L., Stephenson, K., Sumners, D.W., Rehm, K., Schaper, K., & Rottenberg, D.A. (1999). Quasi-Conformally Flat Mapping the Human Cerebellum. MICCAI.

Keywords: MRI, Alzheimer's Disease, Brain Mapping, Circle Packing, FreeSurfer