CSIRO scientists have led a study to develop a machine learning tool that gives medical professionals early warning of a patient's deteriorating condition.
The study was done in collaboration with Brisbane's Princess Alexandra Hospital and Metro South Hospital and Health Service and applied to a test cohort of more than 18,000 patient records.
It found early-warning deterioration alerts can be set to monitor patients for two to eight hours before they are alerted by current clinical measures.
Researchers say medical professionals can use the data found in electronic medical records to predict when a patient's vitals and condition might decline.
"Until now there hasn't been a way to harness all the data in the EMR to predict patient health," the CSIRO's Dr Sankalp Khanna said.
"This new tool has the potential to transform the day-to-day functioning of health systems."
With the data that comes with EMR, this information can be used by to assist medical staff in decisions regarding a patient's health and care.
Until recently, some hospital and patient data was not accessible electronically, which restricted the ability to develop digital tools from it.
"Our scientists hold expertise in transforming data into usable information to help guide clinical choices," Dr Khanna said.
"The new tool also sets out the reasons for the warning, which can guide the choice of intervention.
"Clinical decision support tools such as these are a pre-emptive solution that can provide medical staff with an opportunity to intervene earlier to prevent adverse patient outcomes,."
In the test cohort, the tool had a 100 per cent sensitivity for prediction windows two to eight hours in advance.
This was for patients that were identified at 95 per cent, 85 per cent and 70 per cent risk of deteriorating.
Dr David Cook, an ICU staff specialist at the PA Hospital, said the tool could be implemented at large hospitals to manage unexpected patient deterioration.
"It is done without process duplication, nor does it interfere with established best practice systems which are used to recognise sick and deteriorating ward patients," he said.
A clinical trial is being discussed by CSIRO scientists and partners to find how the alerts work and their best implementation.
