Mobile application of finger tapping task assessment for early diagnosis of Parkinson's disease

Abstract

A mobile application named HLTapper was designed for Android mobile phones to collect and assess real-time alternating finger tapping movement data such as mean tapping speeds, temporal variations of tapping intervals, horizontal tapping variances, and vertical tapping variances, from built-in sensors while a subject performs the finger tapping movement task according to directions given through the application interface. The results of a controlled experiment (40 subjects including nine subjects with Parkinson disease, 11 healthy age-matched subjects, and 20 healthy young subjects) and a discriminant analysis revealed sensitivity of 85.71% and specificity of 91.42%, on an average, which suggests that HLTapper would be useful for early diagnosis and personalised treatment plan adjustments of patients with Parkinson disease.