An algorithm designed by security and aerospace contractor Lockheed Martin was used to successfully provide early detection of sepsis cases in an initial trial at the University of Pennsylvania in Philadelphia.
The algorithm is based on Lockheed Martin's missile defense technology. Missiles, like sepsis, are best detected by quickly and accurately monitoring a large number of constantly changing signals for anomalies.
"Vitals and lab data are, essentially, continuously updated streams of information — just like missile signals detected by our defense systems," said Melanie Lang, Lockheed Martin business development lead within Information Systems & Global Solutions, in a news release. "Lockheed Martin took insights from real-time, streaming sensor data, detecting missiles at mach speeds, and reapplied those same techniques to real-time patient data."
In the trial, 4,500 patients were tested for sepsis using Lockheed Martin's algorithm, which accurately detected sepsis 90 percent of the time while providing false-positives less than 1 percent of the time. By contrast, the current method for sepsis detection used by physicians, the Systemic Inflammatory Response System, detects sepsis 69 percent of the time and will incorrectly diagnosis as many as 65 percent of uninfected patients. The Lockheed Martin algorithm was also able to detect sepsis 14 to 16 hours sooner than the current diagnostic approach.
Lockheed Martin is expanding their data analytics foray into healthcare — the company recently entered into a partnership with the University of Miami and the Miller School of Medicine to explore more ways to harness Lockheed Martin's data analytics capabilities to improve patient care.
More Articles on Big Data:
Illinois, Missouri Hospital Associations Partner to Improve Care Through Data Analytics
UCLA Researchers Use Big Data From Twitter to Locate HIV Outbreaks
Untethered: How Children's Medical Center Dallas Is Improving Care By Sharing Data, Innovations