A Diagnosis Support System for Finger Tapping Movements Using Magnetic Sensors and Probabilistic Neural Networks

This paper proposes a system to support diagnosis for quantitative evaluation of motility function based on finger tapping movements using probabilistic neural networks (PNNs). Finger tapping movements are measured using magnetic sensors and evaluated by computing 11 indices. These indices are standardized based on those of normal subjects, and are then input to PNNs to assess motility function. The subject’s motor ability is probabilistically discriminated to determine whether it is normal or not using a classifier combined with the output of multiple PNNs based on bagging and entropy. This paper reports on evaluation and discrimination experiments performed on finger tapping movements in 33 PD patients and 32 normal elderly subjects. The results showed that the patients could be classified correctly in terms of their impairment status with a high degree of accuracy (average rate: 69 . 3 1 . 93 ± %) using 12 LLGMNs, which was about 5% higher than the results obtained using a single LLGMN.