by Nancy Ryerson
, Staff Writer | November 07, 2013
Doctors at Emory University want to predict the future. Or at least get as close as they can, using real-time patient data streaming from a team-up with IBM and Excel Medical Electronics.
The new system, currently in a research stage, will allow clinicians to better direct the deluge of data each patient in the ICU produces. Excel Medical collects data from physiologic monitors and makes it available to other systems, while IBM Infosphere Streams takes in, analyzes and correlates that information, from up to thousands of sources. Together, the systems will combine to help clinicians make sense of patient data as it happens.
"What used to be just a little trickle of data, with caregivers taking down vital signs on a piece of paper, has become a tidal wave of information that no one human being can keep track of," Dr. Tim Buchman, director of critical care at Emory University Hospital, told DOTmed News. "Every day each of us as caregivers gets up knowing we're going to make dozens of decisions about each patient, and we know they can't all be right. So how do we get to the desired goal of doing the right thing every time?"
He said the research team has looked to other industries, such as weather forecasting and banking, that have managed to part the sea of data to their benefit.
"If you live in Mobile, Alabama and you hear that a hurricane is coming, you can make personalized predictions," said Buchman. "Being able to predict the patient trajectory will allow us to make personalized choices for the patient. For example, we're going to predict that the patient is going to produce 400 milliliters of urine over the next two hours. When the patient only produces 120, this patient has done something unexpected and we can change course."
Though the way the streaming data will be visualized has not yet been determined, Buchman said he knows for certain that no one needs more bleeping alarms in the hospital. He hopes technology can also streamline alerts and help prevent alarm fatigue, a common complaint from nursing staff.
And while the technology is not yet a reliable crystal ball, Buchman said they're starting to be able to see where problems are forming and keep small issues from becoming catastrophes.
"We're not quite at the point where we can predict who's going to have a disaster, but we can see the footprints of the problem forming," said Buchman. "These are common problems, such as patients developing abnormal heart rhythms or accumulating fluid in the lungs."If we can find those things a few minutes or an hour before they become obvious, intervening then and not waiting for the patient's condition to unravel is going to result in better care and take cost out of the system."
Back to HCB News