NeuroANATOP: An Effective Tool to Assess the Progression of Alzheimer's Disease in Human Brain Networks

Document Type

Conference Article

Publication Title

Proceedings - IEEE International Conference on Knowledge Graph, ICKG 2023

Abstract

Alzheimer's disease (AD) is one of the significant neurocognitive disorders, generally occurring among older population that progressively worsens with time initially showing symptoms of mild cognitive impairment (M C I). People suffering from AD usually lose their thinking skills and eventually fail to manage their daily routine tasks because of decline in successive memory functions. There exist well-documented evidences suggesting constant deterioration of communication among anatomical regions of the brain in individuals with Alzheimer's disease (AD) and mild cognitive impairment (M C I). Recent studies also exhibit the differences in topological properties of brain network between the individual patients suffering from AD and MC I compared to healthy controls (HC). In this work, we propose a novel hybrid model NeuroANATOP by employing both the anatomical and the topological properties of a brain network to capture the structural and/or topological changes occurring in networks during the progression from H C to MC I and H C to AD. Given a brain atlas, the Euclidean distance between two regions of interest (ROIs) in the brain is regarded as a measure of anatomical distance, quantifying an anatomical property. On the other hand, the topological similarity, quantifying a topological property, is defined by taking the cosine distance between nodes after embedding the whole real brain network into a d-dimensional vector space. Empirical results over real-brain network data suggest that the proposed hybridization successfully models the topological differences observed across the transition from HC to MCI and HC to AD.

First Page

108

Last Page

116

DOI

10.1109/ICKG59574.2023.00019

Publication Date

1-1-2023

This document is currently not available here.

Share

COinS