Early Parkinson’s Disease Detection from EEG Traces Using Machine Learning Techniques

Parkinson’s disease (PD) affects over 10 million people worldwide. Tremors, stiffness, voice changes, delayed movement, and difficulty walking are typical symptoms of the condition. In the early stage of the disease, these symptoms are difficult to identify and therefore it becomes increasingly important and necessary to be able to predict Parkinson’s and diagnose it as soon as possible. In particular, more studies have focused on the differences found in the electroencephalogram (EEG) of patients with PD. The EEG is used to measure the electrical activity of the brain. The detection of unusual signals is indicative of a pathological condition such as Parkinson’s. This work proposes a new way of researching the diagnosis and surveillance of PD. The results are satisfactory and better when compared with those of other studies conducted on the same data. This indicates that the proposed method can effectively improve early Parkinson’s diagnosis by reducing the time and effort required.