JMIR Publications

ABSTRACT

Background:

Hand tremor typically has a negative impact on a person’s ability to complete many common daily activities. Past work has investigated how to quantify hand tremor with wearable and smartphone sensors mainly under controlled data collection conditions. Solutions for in-the-wild settings remain largely underexplored.

Objective:

Our objective is to monitor and assess hand tremor severity, and to better understand medication effectiveness in a naturalistic environment.

Methods:

Using the Welch method, we generate periodograms of accelerometer data and compute signal features to compare patients with varying degree of PD’s symptoms.

Results:

We introduce and empirically evaluate the Tremor Intensity Parameter (TIP), an accelerometer-based metric to quantify hand tremor severity in Parkinson Disease using smartphones. We report a statistically significant correlation between TIP and self-assessed UPDRS II tremor scores (Kendall rank correlation test: z = 30.521, P= 2.2e-16, tau = 0.5367379, N=11). An analysis of the before and after medication intake conditions identified a significant difference among patients with different levels of rigidity and bradykinesia (Wilcoxon rank sum test, P<.05).

Conclusions:

Our work demonstrates the potential to use smartphones’ inertial sensors as a systematic symptom severity assessment mechanism for monitoring PD symptoms and assessing medication effectiveness remotely. Our smartphone-based monitoring app can be relevant also for other conditions where hand tremor is a prevalent symptom.