PET analysis using features from intensity size zone matrix for group difference between mild cognitive impairment and normal control

Abstract

Alzheimer's disease (AD) is currently one of the important diseases affecting many patients. Unfortunately, the disease is one of the intractable diseases with no easy cure. Neuroimaging is commonly used for the diagnosis of AD. Positron emission tomography (PET) is representative neuroimaging methods for AD. Many studies have been conducted using PET images based on various image processing techniques to assess AD. Texture analysis is a method uses information and various features derived from the gray-level intensity and position of neighborhood voxels in a region of interest (ROI) to accurate describe a group of patients. Representative texture analysis methods include histogram based analysis, gray level cooccurrence matrix (GLCM) based analysis, and intensity size zone matrix (ISZM) based analysis. Here we adopted two features from ISZM to compare mild cognitive impairment (MCI) and normal control (NC) groups in both hemispheres of the hippocampus. Texture analysis was applied to two types of PET imaging, fludeoxyglucose and Pittsburgh compound B. As a result, we found significant differences between MCI and NC groups. © 2017 IEEE.