When the body’s reaction to an infection goes awry, it can result in sepsis, a potentially fatal medical emergency that causes widespread inflammation, organ failure, and even death. To avoid serious complications, sepsis needs to be treated right away.
Canadian researchers have developed a novel portable device and blood test that can identify early sepsis symptoms faster and more accurately than existing diagnostic methods.
The new test, which was described in a recent issue of Nature Communications, has an accuracy rate of over 90% in identifying those who are at high risk of sepsis, which represents a major breakthrough in the diagnosis and treatment of the illness.
“Sepsis accounts for roughly 20 per cent of all global deaths,” explained the lead author Dr. Claudia dos Santos, a critical care physician and scientist at St. Michael’s Hospital. “Our test could be a powerful game changer, allowing physicians to quickly identify and treat patients before they begin to rapidly deteriorate.”
According to the study’s researchers, early detection is especially difficult because early symptoms are not always clear and current diagnostic procedures can take up to eighteen hours and require specialist labs. The chance of dying increases by around 8% for every hour that treatment is postponed.
Researchers from the University of British Columbia (UBC) and Sepset, a biotech company affiliated with UBC, used machine learning to analyze blood samples from more than 3,000 hospitalized patients suspected of having sepsis. They discovered a six-gene expression pattern called “Sepset” that could predict sepsis in 90% of cases, even before a formal diagnosis was made. A 94% accuracy rate in detecting early-stage sepsis in patients whose condition was about to deteriorate was attained by the test in a follow-up validation research that used 248 blood samples using a common hospital technique called RT-PCR (Reverse Transcription Polymerase Chain Reaction).
Machine learning continues to play a vital role in both the treatment and diagnostics of medical conditions, which is likely to see greater integration into the field of medicine in the year ahead.
The National Research Council of Canada (NRC) created the PowerBlade, a small device designed to simplify the administration of the test at the point of service. The gadget swiftly detects sepsis symptoms by following an automatic sequence of actions with just one drop of blood. PowerBlade showed 92% accuracy in identifying patients at high risk for sepsis and 89% accuracy in confirming those who weren’t in studies with 30 patients.