Deep Learning-based Eye-Tracking Analysis for Diagnosis of Alzheimer's Disease Using 3D Comprehensive Visual Stimuli

• DOI: 10.48550/arxiv.2303.06868
• Publication Date: 2023-01-01
• Authors: Fangyu Zuo; Peiguang Jing; Jinglin Sun; Jizhong; Duan Duan; Yong Ji; Yu Liu
• Keywords: Convolution (computer science)

Alzheimer's Disease (AD) causes a continuous decline in memory, thinking, and judgment. Traditional diagnoses are usually based on clinical experience, which is limited by some realistic factors. In this paper, we focus exploiting deep learning techniques to diagnose AD eye-tracking behaviors. Visual attention, as typical behavior, of great value detect cognitive abnormalities patients. To better analyze the differences visual attention between patients normals, first conduct 3D comprehensive task non-invasive system collect heatmaps. We then propose multi-layered comparison convolution neural network (MC-CNN) distinguish normals. MC-CNN, representations heatmaps obtained hierarchical encode eye-movement behaviors, further integrated into distance vector benefit task. Extensive experimental results collected dataset demonstrate that MC-CNN achieves consistent validity classifying normals with data.