Sepsis is a potentially fatal condition caused by the body’s reaction to infection. Left untreated, it causes progressive organ failure and eventually, death.

An estimated 5,000 Australians die of sepsis every year, and the disease has a mortality rate of up to 50%. Early detection is vital, but it can be hard to spot in its early stages.

Clinicians at Western Sydney Local Health District (WSLHD) emergency departments have worked with WSLHD Informatics team to develop an algorithm that detects cases of sepsis in emergency departments and alerts clinicians to so they can commence treatment earlier.

It’s been rolled out in adult emergency departments in WSLHD hospitals over the past year with great success.

“We found the algorithm and alert system improved detection of sepsis by 20% above what the clinician does,” said Professor Naren Gunja, Chief Medical Information Officer at WSLHD.

“We were expecting it to improve detection, but not by that much.”

As a person passes through the emergency department, clinicians record a huge amount of information in their electronic file.

The AI then monitors these files in real-time and alerts clinicians when it detects that someone is starting to show signs of sepsis so that the clinicians can commence treatment.

“The AI is there to support clinicians in their decision-making,” said Professor Gunja. “It can see all the observations and results that are available for a patient – including things the clinician might have missed – and allows them to administer treatment sooner.”

Treatment usually involves antibiotics to take care of the underlying infection and ideally needs to be administered within an hour of diagnosis.

After a successful roll-out in adults in the WSLHD, the team, led by Lead Principal Investigator Dr Amith Shetty, have received a Rapid and Applied Research Grant from Sydney Health Partners to adapt the program for use in children.

“At the moment the system is only applicable for adults, but up to half of sepsis cases are in children,” said Professor Gunja. “By combining patient information from Westmead Hospital and the Children’s Hospital at Westmead, we can create an age-agnostic program that can be applied to all hospitals in NSW.”