Classification of Alzheimer’s Disease and Prediction of Mild Cognitive Impairment Conversion Using Histogram-Based Analysis of Patient-Specific Anatomical Brain Connectivity Networks

Article type: Research Article

Authors: Beheshti, Iman^{a; *} | Maikusa, Norihide^a | Daneshmand, Morteza^b | Matsuda, Hiroshi^a | Demirel, Hasan^c | Anbarjafari, Gholamreza^{b; d} | for the Japanese-Alzheimer’s Disease Neuroimaging Initiative¹

Affiliations: [a] Integrative Brain Imaging Center, National Center of Neurology and Psychiatry, Kodaira, Tokyo, Japan | [b] iCV Research Group, Institute of Technology, University of Tartu, Tartu, Estonia | [c] Department of Electrical and Electronic Engineering, Biomedical Image Processing Group, Eastern Mediterranean University, Famagusta, Mersin, Turkey | [d] Department of Electrical and Electronic Engineering, Hasan Kalyoncu University, Gaziantep, Turkey

Correspondence: [*] Correspondence to: Iman Beheshti, Integrative Brain Imaging Center, National Center of Neurology and Psychiatry, Tokyo, 187-8551, Japan. E-mail: [email protected].

Note: [1] Data used in this article were obtained from Japanese-Alzheimer’s Disease Neuroimaging Initiative (J-ADNI). Access tothe original data of the J-ADNI is available on request from the NBDC Human Database (http://humandbs.biosciencedbc.jp/en/) hosted by the National Bioscience Database Center (NBDC) of the JST.

Abstract: In this study, we investigated the early detection of Alzheimer’s disease (AD) and mild cognitive impairment (MCI) conversion to AD through individual structural connectivity networks using structural magnetic resonance imaging (sMRI) data. In the proposed method, the cortical morphometry of individual gray matter images were used to construct structural connectivity networks. A statistical feature generation approach based on histogram-based feature generation procedure was proposed to represent a statistical-pattern of connectivity networks from a high-dimensional space into low-dimensional feature vectors. The proposed method was evaluated on numerous samples including 61 healthy controls (HC), 42 stable-MCI (sMCI), 45 progressive-MCI (pMCI), and 83 AD subjects at the baseline from the J-ADNI data-set using support vector machine classifier. The proposed method yielded a classification accuracy of 84.17%, 70.38%, and 61.05% in identifying AD/HC, MCIs/HCs, and sMCI/pMCI, respectively. The experimental results show that the proposed method performed in a comparable way to alternative methods using MRI data.

Keywords: Alzheimer’s disease, anatomical connectivity networks, feature extraction, magnetic resonance imaging, mild cognitive impairment

DOI: 10.3233/JAD-161080

Journal: Journal of Alzheimer's Disease, vol. 60, no. 1, pp. 295-304, 2017