Alzheimer's Disease Classification Through Imaging Genetic Data With IGnet

• DOI: 10.3389/fnins.2022.846638
• Publication Date: 2022-03-03T04:49:09Z
• Authors: Jade Xiaoqing Wang; Yimei Li; Xintong Li; Zhao-Hua Lu
• Keywords: Autoencoder; Data set

The application of deep learning techniques to the detection and automated classification Alzheimer's disease (AD) has recently gained considerable attention. rapid progress in neuroimaging sequencing enabled generation large-scale imaging genetic data for AD research. In this study, we developed a approach, IGnet, using both magnetic resonance (MRI) data. proposed approach integrates computer vision (CV) natural language processing (NLP) techniques, with three-dimensional convolutional network (3D CNN) being used handle MRI input Transformer encoder manage sequence input. been applied Disease Neuroimaging Initiative (ADNI) set. Using baseline scans selected single-nucleotide polymorphisms on chromosome 19, it achieved accuracy 83.78% an area under receiver operating characteristic curve (AUC-ROC) 0.924 test results demonstrate great potential multi-disciplinary AI approaches integrate AD.