Early Diagnosis of Alzheimer’s Disease Based on Deep Learning and GWAS

Abstract

Alzheimer’s disease (AD) is a typical irreversible neurodegenerative disease. At present, the pathogenesis of AD remains elusive and the effective treatment of AD is still a challenge for clinicians. Therefore, early diagnosis is of great importance for the development of new drugs to prevent the progression of AD. With the rapid advancement of neuroimaging technology and deep learning, more and more researchers have turned to deep learning to analyze the brain images for early diagnosis of AD. Plus, studies have demonstrated that it is very likely that the genetic makeup of an individual may influence his/her susceptibility to AD traits. Researchers have begun to identify the genetic biomarkers associated to AD and evaluate the effects of genes upon the changes in the structure and function of the brain of AD patients. In this study, an ensemble model of multi-slice classifiers based on convolutional neural network (CNN) was proposed to make an early diagnosis of AD and at the same time to identify the significant brain regions related to AD. The morphological data of these identified brain regions and the genotype were utilized to carry out genome-wide association studies (GWAS) to explore the potential genetic biomarkers of AD.