Unveiling the Potentials of Deep Learning Techniques for Accurate Alzheimer’s Disease Neuro Image Classification

Abstract

This manuscript explores the potential of deep learning strategies, including convolutional neural networks (CNN); recurrent neural network (RNN); long short term memory (LSTM); Bi-directional LTSM (Bi-LSTM), for accurate and automated classification of neuroimages in the diagnosis of Alzheimer’s Disease (AD). The study introduces an optimized deep learning model, Bi-LSTM-AJSO, developed using the artificial jellyfish search optimization (AJSO) algorithm. The study conducts a thorough contrast with machine learning and deep learning techniques, along with optimized Bi-LSTM models. The summarized outcomes illustrate the consistent superiority of Bi-LSTM in accuracy, precision, sensitivity, and F1-score. The proposed Bi-LSTM-AJSO stands out among rest of the optimized Bi-LSTM models and surpasses other machine learning and deep learning-based classifiers. Notably, promising convergence and execution time performances are observed for both Bi-LSTM and Bi-LSTM-AJSO. These findings emphasize the effectiveness of Bi-LSTM models and optimized methods in precise AD neuroimage analysis. The research suggests future avenues, including exploring alternate deep learning architectures, and underscores the significance of independently validating datasets for real-world clinical relevance.