Classification of Alzheimer’s Disease Using Deep Learning based Edge Detection and Fuzzy Neural Network

Abstract

Purpose: Alzheimer's disease is a neurological ailment that gradually decreases a person's capacity for daily tasks, memory, and reasoning.

Methods: This paper presents the use of a deep learning-based edge detection method called holistically-nested edge detection for Alzheimer's disease categorization. The processed images are fed into a fuzzy neural network's convolutional architecture in order to extract features from the previously processed images. To achieve better categorization, a fuzzy interface system is then used. For several picture kinds, this system explores rich feature representation.

Results: The average accuracy using the MRI dataset is determined to be 95.75% when tested against the dataset including several categories of images related to Alzheimer's disease.

Conclusion: The simulation result exemplifies that proposed method can be capable to determine ideal global solutions efficiently with exactly than existing methods.