Development of an EEG-Based Brain Stimulation User Interface for Managing Rigidity in Parkinson's Disease

Parkinson's disease (PD) is a progressive neurological condition that causes rigidity, bradykinesia, and tremors, among other motor symptoms that significantly lower quality of life in patients. Deep brain stimulation (DBS) and levodopa are two popular treatments that frequently don't give patients consistent comfort. The goal of this project is to create a novel EEG-based brain stimulation user interface that is especially intended to control rigidity in Parkinson's disease patients. The suggested method uses high-fidelity electrodes that are Bluetooth-connected to transfer data in real-time while capturing EEG signals from the motor cortical regions. The main goal of signal processing is to extract power spectral information from the frequency bands of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$a$</tex> (alpha), β (beta), and θ (theta). The β band is especially noteworthy because of its strong correlation with stiffness and motor control. The intensity of stiffness is divided into three categories by machine learning models, such as Random Forest (RF), K-Nearest Neighbors (KNN), Decision Tree (DT), and Support Vector Machine (SVM). This classification optimizes treatment settings for individualized therapy by facilitating adaptive brain stimulation and real-time therapeutic decision-making. The closed-loop system continually analyses EEG readings, categorizes the degree of rigidity, and modifies stimulation parameters in real time. Interviews with 10 physicians, including neurologists and neurosurgeons, were done in order to validate the method. High suggestibility was revealed by the response, provided that safety precautions and ethical clearances are properly followed. The goal of this comprehensive approach is to improve the precision and effectiveness of treating rigidity in Parkinson's disease (PD) by addressing the variability in present treatments. Through the utilization of forefront machine learning techniques and live EEG monitoring, the suggested interface presents a viable, non-invasive method for enhancing patient outcomes and quality of life.