 Parkinson's disease, PD, is a neurodegenerative disorder which can be accurately diagnosed using a combination of upper limb movement patterns and sensors. A user-friendly auxiliary diagnostic system was developed based on collected data from 100 subjects, including 50 PD patients and 50 healthy subjects. The system consists of wearable sensors, a host computer, and a graphic interface, GUI, which uses a genetic algorithm optimized random forest classifier to distinguish between PD and normal states. The system was tested with 50 trials of leave one out cross validation, achieving an overall accuracy of 94.4%. Additionally, the system was evaluated across various upper limb movement tasks and with varying numbers of sensors. Results showed that the task with only alternating hand movements had satisfactory classification accuracy, while sensors on both wrists performed better than one sensor on a single wrist. The system was demonstrated to neurologists with a deployed GUI, demonstrating its potential use in the auxiliary diagnosis of PD. This article was authored by Minchin, Junfeng Sun, Fesu, and others. We are article.tv, links in the description below.