Patients in an intensive care unit typically are surrounded by machines and monitors measuring heart rate, lung capacity and other vital signs. Clinicians must take in and process multiple streams of data, which is typically analyzed retrospectively.
At the Emory University School of Medicine, a team headed by Dr. Tim Buchman, a professor of surgery and director of the school's Center for Critical Care, is using IBM's streaming analytics platform along with data aggregation software from Excel Medical Electronics to collect and analyze more than 1,000 real-time data points per patient per second from between four and six independent data streams.
Although Emory's "ICU of the Future" project is still in the research phase, it promises to help physicians recognize potentially life-threatening scenarios and intervene proactively.
Jack Kearse / Emory University Emory University’s big data team: Gari Clifford, Myffy Hopkins, Terry Willey (seated) and Sharath Cholleti.
"By displaying heart rate and oxygen saturation rate simultaneously and updating it as frequently as allowable, we can detect life-threatening clinical events that folks at the bedside are missing," says Buchman, who is also a practicing physician. Additionally, clinicians would be able to assess current patient data in relation to historical data from thousands of previous patients.
Essentially, the system would identify archetypes of patient types by looking at the outcomes and trajectories of thousands of other patients. "It would be a quantum shift in caring for patients more rapidly and efficiently, and in a more targeted manner," says Dr. Tommy Thomas, a critical care physician at Emory.
For now, "this is all going on in a research study, but we're quite confident that caretakers will get earlier warnings and prevent significant disability and sometimes even death once exposed to these novel views of data," he adds.
Full implementation will require FDA approval, and Buchman estimates that that remains "years away."