Title :
Study on texture characteristics of hippocampus in MR images of patients with Alzheimer´s Disease
Author :
Zhou, Xiaoxia ; Liu, Zhuo ; Zhou, Zhen ; Xia, Hong
Author_Institution :
Sch. of Biomed. Eng., Capital Med. Univ., Beijing, China
Abstract :
Alzheimer´s disease (AD) is the most common neurodegenerative disease in individuals older than 65 years of age. Nowadays, the imaging study of AD mostly focuses on morphologic measurement of the early predilection sites such as hippocampus. However, there is still lack of a specific parameter for early diagnosis. In this study, texture analysis was performed on MR images of three groups including 18 AD patients, 18 elderly controls (EC) and 18 young controls (YC) to extract texture characteristics from gray level co-occurrence matrix (GLCM) and run-length matrix (RLM). Then the correlation between texture features and hippocampal volumes was tested. The results showed that there were significant differences on three texture parameters of hippocampus between AD and EC group (p <;; 0.05), namely 1) sum average (SA); 2) difference variance (DV); 3)grey level nonuniformity (GLN), as well as the volumes of each lateral hippocampus between AD and EC group (p <;; 0.05). Our research also demonstrated that there were significant correlations between the texture features of one sided hippocampus and the corresponding hippocampal volumes respectively (p <;; 0.05), but some texture parameters between EC and YC group had no significant differences. These results indicated that the hippocampal changes with increasing age were essentially different from the pathological changes with AD. This study suggested that the texture characteristics may reflect the hippocampal underlying pathological changes in the early stage of AD, which could be used as an auxiliary means for early diagnosis of AD.
Keywords :
biomedical MRI; brain; diseases; feature extraction; image texture; medical image processing; neurophysiology; Alzheimer Disease; MR images; difference variance; early diagnosis; feature extraction; gray level cooccurrence matrix; grey level nonuniformity; hippocampus; morphologic measurement; neurodegenerative disease; run length matrix; sum average; texture characteristics; Aging; Alzheimer´s disease; Correlation; Hippocampus; Magnetic resonance imaging; Pathology; Volume measurement; Alzheimer´s disease; Hippocampus; MRI; Texture analysis;
Conference_Titel :
Biomedical Engineering and Informatics (BMEI), 2010 3rd International Conference on
Conference_Location :
Yantai
Print_ISBN :
978-1-4244-6495-1
DOI :
10.1109/BMEI.2010.5640016