A Patch Based 3D CNN Approach for Diagnosing Early Stages of Alzheimer's Disease by Applying OBL-WOA Algorithm

• Published in: 2023 8th International Conference on Cloud Computing and Big Data Analytics (ICCCBDA) pp. 455 - 459
• Main Authors: Kumari, Rashmi, Goel, Shivani, Das, Subhranil
• Format: Conference Proceeding
• Language: English
• Published: IEEE 26.04.2023
• Subjects: Alzheimer Disease; Classification algorithms; Convolutional Neural Network; Convolutional neural networks; Deep learning; Magnetic resonance imaging; Medical services; Three-dimensional displays; Whale Optimization; Whale optimization algorithms
• ISSN: 2832-3734
• DOI: 10.1109/ICCCBDA56900.2023.10154706

Alzheimer's disease (AD) is one of the most degraded neurodegenerative brain disorders, with no treatments involved. Understanding the early stages of Alzheimer's disease would be necessary for treating the disease and preventing further Tumour Necrosis Factor α (TNF-α) degeneration cells. Previous studies demonstrated the application of deep learning techniques for distinguishing AD from Normal Control (NC) by applying T1-weighted MRI images. This paper proposes a novel Patch Based Convolutional Neural Network (PB-CNN) network for classifying three binary classifications. A new optimization technique, the Opposition Based Learning- Whale Optimization Algorithm (OBL-WOA), has been proposed to update the weights of the proposed PB-CNN Network. 326 ADNI subjects are investigated for the feasibility of the proposed optimization technique, where the highest classification accuracy is achieved when compared with other state-of-the-art techniques. Moreover, the proposed technique could assist doctors in diagnosing the early stages of AD.