A Comparative Atlas-Based Recognition of Mild Cognitive Impairment With Voxel-Based Morphometry.

Zhuqing Long,Bo Li,Zihao Li,Jinchang Huang,Hongwen Chen,Bin Jing,Zuojia Li

doi:10.3389/fnins.2018.00916

Abstract

An accurate and reliable brain partition atlas is vital to quantitatively investigate the structural and functional abnormalities in mild cognitive impairment (MCI), generally considered to be a prodromal phase of Alzheimer’s disease. In this paper, we proposed an automated structural classification method to identify MCI from healthy controls (HC) and investigated whether the classification performance was dependent on the brain parcellation schemes, including Automated Anatomical Labeling (AAL-90) atlas, Brainnetome (BN-246) atlas, and AAL-1024 atlas. In detail, structural magnetic resonance imaging (sMRI) data of 69 MCI patients and 63 HC matched well on gender, age, and education level were collected and analyzed with voxel-based morphometry method first, then the volume features of every region of interest (ROI) belonging to the above-mentioned three atlases were calculated and compared between MCI and HC groups, respectively. At last, the abnormal volume features were selected as the classification features for a proposed support vector machine based identification method. After the leave-one-out cross-validation to estimate the classification performance, our results reported accuracies of 83, 92, and 89% with AAL-90, BN-246, and AAL-1024 atlas, respectively, suggesting that future studies should pay more attention to the selection of brain partition schemes in the atlas-based studies. Furthermore, the consistent atrophic brain regions among three atlases were predominately located at bilateral hippocampus, bilateral parahippocampal, bilateral amygdala, bilateral cingulate gyrus, left angular gyrus, right superior frontal gyrus, right middle frontal gyrus, left inferior frontal gyrus, and left precentral gyrus.

Highlights

Mild cognitive impairment (MCI), which represents the transition state between normal aging and the early changes related to Alzheimer’s disease (AD) (Han et al, 2011; Wang et al, 2015; Khazaee et al, 2016, 2017), is characterized by intellectual and cognitive deficits, memory complaints, and behavioral disturbances (Zhang et al, 2012; Beheshti et al, 2017), and generally regarded as aAtlas-Based mild cognitive impairment (MCI) Identification With voxel-based morphometry (VBM) prodromal phase of AD (Long et al, 2018)
This study focused on comparing the classification performance of identifying MCI patients from healthy controls (HC) subjects with VBM under three widely used brain atlases, and found that the performance varied in different brain atlases
A radial basis function (RBF) kernel function that could deal with the nonlinear relationship between features and labels was adopted to improve the classification performance (Hsu et al, 2003)

Summary

Introduction

Mild cognitive impairment (MCI), which represents the transition state between normal aging and the early changes related to Alzheimer’s disease (AD) (Han et al, 2011; Wang et al, 2015; Khazaee et al, 2016, 2017), is characterized by intellectual and cognitive deficits, memory complaints, and behavioral disturbances (Zhang et al, 2012; Beheshti et al, 2017), and generally regarded as aAtlas-Based MCI Identification With VBM prodromal phase of AD (Long et al, 2018). Structural magnetic resonance imaging (sMRI) has been prevalently utilized to characterize differences in shape and neuroanatomical configuration in MCI and AD because it could provide visualization of the macroscopic tissue atrophy caused by the cellular changes underlying MCI and AD (Desikan et al, 2009). Some studies employed sMRI data to identify MCI or AD from healthy controls (HC) by extracting structural characteristics such as voxel-wise volume (Fan et al, 2007; Davatzikos et al, 2008a,b; Klöppel et al, 2008; Magnin et al, 2009; Beheshti and Demirel, 2016) and vertex-based cortical thickness (Lerch et al, 2008; Eskildsen et al, 2013; Dimitriadis et al, 2018), and the classifying accuracies varied largely from 58% to 100%, which indicated that the discriminative diagnoses of MCI and AD with sMRI data need to be continued

Methods

Results

Conclusion