You’ve been there. In the doctor’s office that is. You’re not feeling well and you want to tell the doc all about it, but he wants to ask you questions like: How would you rate the pain on a scale of 1 to 10? When did this start? How long does it last? How would you rate your sleep 1 to 10?
You answer the questions with what is, at best, a guess. And the doctor makes assessments based upon your answers. But is what he calls an “8” the same thing you call an 8? And what does “sleeping well” actually mean? (And, yes, we know there’s information like heart rate, blood pressure, lab work data to consider, but we’re putting that aside for the moment.)
Now forget about yourself and think of a Parkinson’s patient. Michael J. Fox or Intel’s Andy Grove may be the ones we “know” best, unless there’s someone in our personal lives who has been affected. Their doctors probably include some physical tests in their visits, like asking them to touch their fingertips to their noses or to walk a straight line by placing one foot in front of the other.
Patient performance on activities like these varies. We all have good days and bad. And treatments and research, especially for those who deal with hard-to-manage diseases are, on a large part, based on what a doctor observes during an office visit, what data a patient provides at a specific point in time and what existing medical research suggests.
This isn’t bad medicine. It’s everyone doing their best given the available tools.
Up until now, that is.
Changing the Game
Today the Internet of Things (IoT), big data, data scientists, medical professionals and their patients have an opportunity to change the game, and Intel and Michael J. Fox Foundation for Parkinson’s Research (MJFF) are going for it.
This afternoon Diane Bryant, senior vice president and general manager of Intel's Data Center Group, and Todd Sherer, CEO of The Michael J. Fox Foundation announced a collaboration aimed at improving research and treatment for Parkinson's disease.
It consists of a new big data analytics platform that detects patterns in data collected from wearable devices used to monitor symptoms in Parkinson’s patients from which medical researchers and other professionals can glean insights.
With the advent of wearable devices and always-on data transmitters, big advances can be made from when, where and how often data is collected. Bryant says that wearables can unobtrusively gather and transmit objective, experiential data in real time, 24 hours a day, seven days a week.
This means that instead of working with a very small number of data points, such as those collected during an office visit or via patient diaries (which are, by their very nature, subjective and not always used) researchers can now access and analyze hundreds of readings per second from thousands of patients to attain a critical mass of data to detect patterns and make new discoveries.
“Collecting the data is not the challenge,” according to Sherer. It can be as simple as having a patient wear a data-gathering watch and transmitting the data to the cloud via mobile phone.
How much data are we talking about? Up to 300 observations per second per device, said Bryant.
Collecting, storing and managing the data isn’t constrained either. Intel products certainly have the muscle, and Cloudera’s Hadoop distro, which is the platform the project runs on, on Amazon’s Cloud, can handle whatever you throw at it; volume, velocity, variety and variability aren’t issues.
And it’s not just patient IoT data that will be leveraged over the long term. Data such as patient, genome, clinical trial data, and other kinds of medical research data could be incorporated as well.
More than Data
But it’s not just the data that is the differentiator.
It’s what you can do with the data and who uses it that matters as well.
Intel and MJFF have planned this perfectly in that they’ve invited both data scientists and medical researchers to participate in the undertaking.
Bryant referred to the project as a three legged stool made up of IT, data scientists and medical researchers who will work together to provide new, important research about Parkinson’s.
And that’s just the tip of the iceberg.
Though, for the moment, Intel’s focus is on working with MJFF on Parkinson’s, much will be learned from not only from this initial approach, but also from the possible application of other advanced techniques such as machine learning and graph analytics. They might deliver more accurate predictive insight to detect change in disease symptoms as well as to measure the efficacy of new drugs and assisting physicians with prognostic decisions.
Techniques used here may be able to be applied to other areas of medical research as well.
It’s a big day for big data, for medical research and for those who suffer from or are affected by Parkinson’s.
- Hey Cloudera & MapR: Open Data Platform is the Real Deal
- Discussion Point: Why Do Intranets Fail?
- A Look at Gartner's Data Management Analytics Leaders
- The Sticking Point with Social Collaboration Tools
- 3 Ways Marketing Automation Boosts Business Efficiency
- Is There a Future in Content Marketing?
- 3 Vendors Lead the Wave for Big Data Predictive Analytics