The department of neurosurgery at the University of California, Los Angeles, is using big data analytics software from IBM and Excel Medical Electronics to predict rising brain pressure in patients with traumatic brain injuries.
In the collaboration announced March 13, the public research university will use IBM InfoSphere Streams and EME BedMasterEX to create data models to predict when brain injuries might become life-threatening for patients at Ronald Reagan UCLA Medical Center.
About 1.7 million people per year in the United States suffer a traumatic brain injury, according to the Centers for Disease Control and Prevention.
Although researchers have monitored and predicted changes in brain pressure for patients for years, being able to analyze the patient data flowing in real time has been a challenge, according to Charlie Schick, IBM director of big data health care and life sciences.
“Doctors need to be able to measure intracranial pressure, and up to now, most of the time it’s after the fact reading [data] off the machine,” Schick told eWEEK.
People can monitor data using the software and alert doctors when intracranial pressures change or become dangerous, he said.
Being able to calculate vital-sign data in real time is a “big change in how doctors are dealing with intracranial pressure normally in the clinic,” Schick said.
UCLA’s department of neurosurgery launched the project after being awarded a $1.2 million grant from the National Institute of Neurological Disorders and Stroke to study intracranial pressure and create an alarm system that predicts brain injuries.
Before, doctors would record vital signs and calculate the risks offline with the tools they had, leading to a delay. “If they were delayed, they were not catching these intracranial pressure changes until they got to the point where the patient was at risk of being in a critical state,” Schick said.
IBM InfoSphere Streams is an application that was developed for the military to fight cyber-attacks, according to Schick. When it’s used for cyber-security, large amounts of network traffic data can provide clues to suspicious activity online, and now IBM has provisioned the software for health care to read data coming off patient monitoring systems in real time.
“InfoSphere Streams does all the scoring on the fly,” said Schick. “Looking at data immediately is more important than storing it somewhere.”
Vital signs show subtle changes in a patient’s pulse, blood and intracranial pressure, heart activity and respiration. Collecting this physiological data in real time allows doctors to know when a brain injury crosses a threshold into a risk of brain damage or death.
“The field of big data analytics is evolving to include new kinds of data from sources such as medical monitors, giving us insights into patients that weren’t previously possible,” Dr. Martin Kohn, chief medical scientist at IBM Research, said in a statement. “We believe that UCLA’s promising research may one day transform the way that doctors and nurses interact with patients inside the neuro-intensive care unit.”
InfoSphere Streams collects the data from EME BedMasterEX, a tool that aggregates streams of information from the vital sign monitors. “Then the calculation is done within InfoSphere Streams and reported back as an alert or as something to stream,” Schick said.
The IBM software creates a data model to allow researchers to have a fuller view of the symptoms causing traumatic brain injuries, Schick said.
“Through its research, UCLA is changing the way that analytics may eventually be used inside hospitals, and how we think about using data to improve patient outcomes,” John Hoffman, president of EME, said in a statement. “We hope that teaming Excel Medical Electronics and IBM analytics technologies with UCLA’s expertise will lead to new ways to save lives.”
In addition to brain injuries, data modeling through InfoSphere Streams could be used to predict heart problems or seizures for epileptic patients, Schick said.
“I think we’re just at that opening edge of what we can do here,” he said.