© 2020 MJH Life Sciences™ and Neurology Live. All rights reserved.
Home Use Smartphone App for Parkinson Disease
New technology may help patients and providers assess changes in severity of Parkinson disease symptoms more objectively.
Researchers have used machine learning to develop a smartphone app that can rate severity of symptoms at home. Results were published online in JAMA Neurology. The study is one of the largest to evaluate how smartphones can be used to assess changes in Parkinson disease (PD) symptoms over time. “The mPDS [mobile PD disease scores] is a novel measure that provides rapid, remote, frequent, and objective assessment of PD symptom severity on widely available smartphones,” wrote first author Andong Zhan, MS, of John Hopkins University, Baltimore, MD.
Source: Zhan A, Mohan S, Tarolli C, et al. Using smartphones and machine learning to quantify Parkinson disease severity: the mobile Parkinson disease score. JAMA Neurol. March 2018; Epub ahead of print.
Evaluation of PD Symptoms. PD symptoms are most often evaluated clinically. Current methods tend to be subjective and can vary based on the person doing the assessment. Developing a smartphone based method would allow for more frequent assessments at home and may be more objective.
To develop the smartphone method, researchers used machine learning to design an Android app called Hopkins PD. The app assesses 5 activities: voice, finger tapping, gait, balance, and reaction time. The app also has a function for medication reporting. Then they created the mobile PD scores (mPDS) to objectively assess each activity. The mPDS is rated on a scale from 0 to 100, with higher scores indicating greater symptom severity.
The Observational Study. The researchers tested the app in an observational study of 129 patients with PD (95.3% white, 43% women). Twenty-three patients with PD and 17 without PD also completed standard clinical PD assessments. The study took place between October 2016 and March 2017. The analysis included 6148 smartphone activity assessments.
Results showed that mPDS ratings correlated significantly with the Movement Disorder Society Unified Parkinson Disease’s Rating Scale (MDS-UPDRS) total (P < .001), which is commonly used to follow change in PD symptoms over time. The mPDS also correlated significantly with the MDS-UPDRS part 3 (clinician-scored motor evaluation) (P < .001); the MDS-UPDRS part 5 (stages severity of PD) (P < .001), and the Timed Up and Go test (measures mobility and balance) (P = .002).
Further assessments showed that the mPDS was able to capture symptom improvements in response to dopaminergic medication, with a mean improvement of 16.3 points (p=0.01). Moreover, the mPDS captured more fluctuations in symptom throughout the day than standard assessments
While the results are encouraging, the authors mentioned that mPDS should remain complementary to standard PD assessments. “Further validation of the mPDS in a larger sample with patient-relevant anchors is needed. New iterations of the application for Android and iOS smartphones will expand participation and Include additional features and functionality that could provide new insights into PD.” Study participants were mostly white, college educated Android owners, so the results may not apply to a more diverse population.
The new Android app, Hopkins PD, may help patients and providers assess changes in severity of PD symptoms more objectively at home. The symptom rating scale, mPDS, correlated significantly with standard rating scales of PD and captured symptom improvements in response to dopaminergic medication. Additional studies are needed with larger, more diverse populations, as well as to add new functions and adapt the app for iOS.