Fuzzy RNN Model-Based Classification of Alzheimer’s Disease and Dementia Using Brain EEG Signals

Alzheimer’s Disease (AD) and Dementia represent critical healthcare challenges worldwide, underscoring the need for reliable and early diagnostic tools. This study presents an innovative classification framework that combines the strengths of fuzzy logic and Recurrent Neural Networks (RNNs) to enhance the detection of AD and dementia. Leveraging Electroencephalogram (EEG) signals, a non-invasive and informative modality reflecting cognitive function, we propose a Fuzzy RNN model that integrates the sequential learning capabilities of RNNs with the interpretability and uncertainty handling of fuzzy logic. The model is trained on a comprehensive dataset comprising EEG recordings from individuals with Alzheimer’s Disease, various types of dementia, and healthy controls. Evaluation metrics, including sensitivity (96.49%–97.19%), specificity (96.72%–97.85%), and accuracy (97.82%–98.82%), highlight the model’s robustness in differentiating between diagnostic categories. Comparative analyses against traditional classifiers demonstrate the Fuzzy RNN’s superior ability to capture intricate temporal patterns in EEG signals, establishing its potential as a reliable tool for early and accurate dementia diagnosis.