Alzheimer's Disease Detection Using MRI Analysis with Recurrent Deep Networks

Alzheimer’s disease (AD) is a progressive neurodegenerative disorder that severely affectscognitive function, making early and accurate diagnosis critical for effective treatment. In this study,we propose a deep learning-based method utilizing recurrent neural networks (RNNs) to analyze MRIimages for the early detection of Alzheimer’s disease. Our approach leverages the temporal and spatialdependencies of MRI data, which are key to identifying subtle changes in brain structure associatedwith disease progression. By using a combination of long short-term memory (LSTM) networks andconvolutional neural networks (CNNs), we extract high-level features from MRI sequences and modelthe temporal dynamics of structural changes over time.We evaluate the proposed method on a large dataset of MRI scans, comparing it with traditional machinelearning methods and other deep learning architectures. Our results show that the RNN-based approachachieves superior performance in distinguishing between normal, mild cognitive impairment (MCI),and Alzheimer’s cases. The model demonstrates high accuracy, sensitivity, and specificity,outperforming state-of-the-art methods for MRI-based AD detection. These findings suggest thatincorporating temporal information through recurrent networks can significantly enhance the accuracyof Alzheimer’s diagnosis, offering a promising tool for early intervention and patient management.

کلیدواژه ها:

Alzheimer’s Disease ، Recurrent Neural Networks ، Long Short-Term Memory ، Deep Learning ، MRI Analysis