NEW YORK — Intel is collaborating with the Michael J. Fox Foundation (MJFF) to use big data analytics in research and treatment for Parkinson's disease. A multi-phase study used the specially created analytics platform to detect patterns in participant data collected from wearable technologies used to monitor symptoms.
In conjunction with New York’s Mount Sinai Hospital and Tel Aviv Medical Center, Intel and MJFF used an undisclosed smartwatch tethered to a smartphone to track physiological symptoms such as tremors from 16 Parkinson's patients and nine control volunteers. Participants wore the devices during two clinic visits and at home continuously over four days as an alternative to traditional tracking methods. Data was then correlated with clinical observations and patient diaries to gauge the device’s accuracy and develop algorithms to measure disease progression.
“We’ve gathered 300 events per second, or 1 GB of data per patient per day. That’s as much data per month as the entire Library of Congress,” said Diane Bryant, senior vice president and general manager of Intel’s Data Center Group, at a web conference. She believes big data will play a significant role in all industries but especially healthcare. "The life impacting opportunity created by harnessing and mining the data is limitless.”
Intel connected wearables to the cloud, using real time analytics to monitor Parkinson's patients. Source: Intel/MJFF
“I do think there’s a lot of opportunity in a disease like Parkinson ’s disease because there’s a broad array of symptoms -- from motor to non-motor to repertory to sleep,” MJFF CEO Todd Sherer said during a web conference.
The data platform is deployed on an Apache Hadoop cloud infrastructure and Intel Xeon-based servers, with applications to process and detect data changes in real time. While Intel did not provide details on the hardware inside the wearable, Bryant said each had embedded encryption to protect the sensitive data.
“Intel's work with the Michael J. Fox Foundation offers insight into the company's broader strategies in big data, wearable tech, and IOT,” Charles King, principal analyst with Pund-IT, told EE Times. “Intel has also made major acquisitions in [software and embedded systems], chief among them being the 2009 acquisition of Wind River, the leading vendor in embedded system/solution software. I believe the company... should play a major role in pushing them forward.”
Bryant and Sherer said the study was successful and that both organizations hope to expand the wearable analytics program into other areas of research. The system is being built to be agnostic to different opportunities in wearable sensors, Sherer added.
“So much of that platform is extensible to other diseases, other scientific discoveries. The key here is getting proof points or blueprints to replicate this success in other diseases or across healthcare,” Bryant said. “There are potential savings at $450 billion in the US alone from health analytics, through cutting trial times or even using analytics for fraud detection.”
In the near future, the platform could store other types of data such as patient, genome, and clinical trial data, or enable machine learning and graph analytics to deliver more accurate predictive models. Intel and MJFF will begin another study, headquartered in Boston, later this year.
— Jessica Lipsky, Associate Editor, EE Times