Objective: We developed an application of smartphone tapper (SmpT) and sought to determine whether SmpT is applicable for clinical purpose by comparing SmpT parameters to current gold standard methods. We also investigate whether PD can be discriminated from controls using SmpT parameters.

Background: Smartphone applications have been recently developed to use not only for the diagnosis of Parkinson’s disease (PD) but also for monitoring of patients’ status. However, validation of smartphone application in the evaluation of motor dysfunction in PD has not been systematically studied in larger cohort.

Methods: A total of 57 PD patients (6 drug-naïve PD, Hoehn & Yahr stage I – III) and 87 controls were recruited. Timed tapping test was performed with the SmpT and mechanical tapper (MeT). The SmpT application consists of two rectangles of 30 by 45 mm, separated by 15 mm. Subjects were asked to tap each side of rectangles alternatively at the fastest speed for ten seconds. Parameters obtained and/or calculated from SmpT included number of tapping, inter-tap distance, inter-tap dwelling time, total distance of a finger movement tapping speed of each movement of a finger, and tapping errors (tapping outside of squares). Timed tapping test with MeT was performed according to CAPSIT protocol where patients repeatedly tap one tapper at the fastest speed for 10 seconds and two tappers alternatively for 20 seconds. The same trial of tapping task was repeated three times for each hand.

Results: Parameters such as mean number of correct tapping, total distance of finger movement, inter-tap distance, mean dwelling time were significantly different between PD and controls. Mean tapping number assessed by SmpT was significantly correlated with mUPDRS and its bradykinesia subscores as well as mean tapping number obtained with MeT. Multivariate analysis with age and gender as covariates showed that PD can be discriminated from control using SmpT parameters such as total distance of finger movement or dwelling time as predictive variable. ROC analysis of two models showed AUC for dwelling time and total distance 0.87 (95% CI 0.82-0.93) and 0.92 (95% CI 0.88-0.96), respectively.

Conclusions: Our results suggest that smartphone tapping application is comparable to conventional methods assessing motor dysfunction in PD and may be useful in predicting PD in population.

« Back to 2016 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/a-validation-study-of-a-smartphone-based-finger-tapping-application-for-quantitative-assessment-of-bradykinesia-in-parkinsons-disease/