Early Detection of Neurological Disorders Using Frequency Domain Features and Machine Learning

This study explores an innovative approach to Neurological disease detection using machine learning techniques, focusing on frequency domain feature extraction methods. There are different types of Neurological disease. This paper focuses on the early detection of Alzheimer's disease, which is crucial for effective treatment and management. The study investigates three prominent machine learning models: Random Forest (RF), Support Vector Machine (SVM), and K-Nearest Neighbors (KNN). Among the three models, RF demonstrated exceptional performance, achieving an accuracy of 90.08%, sensitivity of 79.51%, and specificity of 97.04%. The high specificity indicates the model's robust ability to correctly identify non- Alzheimer's cases while maintaining strong overall accuracy and sensitivity. Although competitive, the SVM and K-NN models showed relatively less impressive results compared to Random Forest. These findings suggest that the combination of frequency domain feature extraction and the Random Forest algorithm offers a promising approach for Alzheimer's disease detection. The model's high specificity is particularly noteworthy, as it could reduce false positives in clinical settings, leading to more accurate diagnoses. This research contributes to the growing field of AI-assisted diagnostic tools for neurodegenerative diseases and highlights the potential of machine learning in enhancing early detection capabilities for Alzheimer's disease, potentially improving patient outcomes through timely intervention and treatment