Gray level co-occurrence matrix and extreme learning machine for Alzheimer's disease diagnosis

Abstract

Alzheimer's disease (AD) is a chronic neurodegenerative disease, which is one of the biggest challenges in geriatrics. The global incidence of Alzheimer's disease has been on the rise all the time. Since the cause of Alzheimer's disease is unknown and there is no cure at present, early diagnosis of AD is particularly important. Computer researchers have developed various early detection methods based on machine learning and computer vision. In this study, we effectively combined the gray level co-occurrence matrix and the extreme learning machine to carry out automatic diagnosis of AD, used the gray level co-occurrence matrix to extract texture features, combined with the extreme learning machine to classify AD images, so as to obtain higher training speed and better generalization performance. The results of 10×10-fold cross validation experiments show that our method is superior to several advanced methods. Therefore, this method is effective in the detection of Alzheimer's disease, this method can effectively improve the classification accuracy, and also has good generalization.

Keywords

• Alzheimer's disease
• Detection
• Gray level co-occurrence matrix
• Extreme learning machine