Sage Bionetworks launches the Parkinson’s Disease Digital Biomarker DREAM Challenge. This is the first in a series of open, crowd-sourced analytical projects sponsored by Sage Bionetworks and designed to help researchers identify ways to use smartphones and remote sensing devices to monitor health and disease. The challenge will aim to use sensors to identify aspects of Parkinson’s disease (PD) severity. Funding for the challenge has been provided by the Robert Wood Johnson Foundation (RWJF) and The Michael J. Fox Foundation for Parkinson’s Research (MJFF).
The project takes an open, crowd-source approach to address the first step in analysis of sensor data – feature engineering, or the conversion of raw sensor data into analysis-ready data. By engaging a wide community, the challenge will seek to identify the landscape of possibilities to process sensor data for use in health studies.
The challenge will focus on identifying markers of PD severity from sensor data. An estimated five million people worldwide are living with Parkinson’s, a neurodegenerative disorder that can cause tremors, gait issues, speech problems, and interfere with memory. These symptoms can change with disease progression, medical treatment, and some lifestyle choices. The data used in this challenge include accelerometer, magnetometer, and gyroscope data collected in two separate studies of Parkinson’s disease patients.
One of the datasets used in the challenge was collected from research participants through mPower, a patient-centered, iPhone app-based study of symptom variation in PD. mPower, which launched in 2015 with support from the Robert Wood Johnson Foundation, was one of the first studies to use the ResearchKit framework developed by Apple for administration of research studies through iOS. More than 15,000 individuals have enrolled in this study over the past two years and almost 6,000 performed a remote test for gait and balance. “Remote sensor data capture can be used to objectively detect fluctuations in symptoms within a patient such as in response to medication,” said Larsson Omberg, VP of systems biology at Sage Bionetworks. “But we are not yet very good at detecting differences between PD and non-PD participants. A major contributing factor is the naïve approaches to feature engineering that are currently being used.”
The challenge will also use data collected during activities from the MJFF-sponsored Levodopa Response Trial. In this trial, individuals were monitored with three to eight accelerometer sensors while they performed a variety of motor and ethnographically valid tasks. The trial was spread over two in-clinic days and two at-home days. In clinical visits, participants performed 18 tasks, six times each. During home visits, continuous, raw data was captured. “The Michael J. Fox Foundation supports the use of novel, digital technologies to speed Parkinson’s research,” says Mark Frasier, PhD, senior vice president of research programs at MJFF. “Inviting analysis of the Levodopa Response Trial sensor data in the challenge can further engage leaders in the field to advance computational science toward new treatments and a cure for Parkinson’s disease.” These data will be made available at the close of this challenge for researchers to analyze.
The results of this study, which are expected in the fall, will provide best practices and tools to advance the development of PD digital biomarkers, as well as to advance the mobile health community at large. The winning researcher teams with the best methods for processing sensor data will share a $25,000 prize from The Michael J. Fox Foundation.
For additional information, go to https://www.synapse.org/DigitalBiomarkerChallenge