This study investigates the use of machine learning to differentiate between Parkinson’s patients and healthy individuals using functional near infrared spectroscopy (fNIRS) signals collected during walking. fNIRS data, which monitor brain activity, were acquired from both groups to identify distinctive patterns. Using machine learning techniques, the model was trained to discriminate the two classes, with the goal of improving early and non-invasive diagnosis of Parkinson’s disease and the method delivered an accuracy of 72.00%. The results demonstrate the effectiveness of the method in accurately distinguishing patients from healthy individuals from cortical activity during walking.