Using AI To Spot When Patients Condition Deteriorates

Machine learning’s application in healthcare is commonly used to provide earlier diagnoses than would otherwise be the case.  New research from the University of Oxford highlights how this can even be the case when the timeframe is pretty short.

The researchers have developed an algorithm that they believe can improve the ability of clinicians to spot when patients in hospital are deteriorating to such an extent as to require intensive care.

“The HAVEN machine learning algorithm, using electronic patient data collected routinely by most NHS hospitals, has the potential to substantially improve our ability to detect patients who require ICU, and those for whom a timely intervention is likely to change their outcome, so enhancing the National Early Warning Score (NEWS) system currently in use across the health service,” the researchers say.

Vital signs

The new system combines a number of the patients’ vital signs, including their heart rate, temperature, and blood pressure, along with their comorbidities, frailty, and blood test results to produce a unified risk score.  The researchers believe this gives a precise indication of whether the patients’ health is deteriorating or stable.

The scale of the problem is significant, with over 60,000 people per year seeing their condition deteriorate in a British hospital to the extent that they’re admitted to an intensive care unit.

While hospitals have developed warning systems to try and improve their ability to detect people at risk of such deterioration, many of these systems are based on abnormalities in the vital signs of each patient.

“Late recognition of patient deterioration in hospital is associated with worse outcomes, including higher mortality. Despite the widespread introduction of early warning score systems, which are based on vital signs, deterioration still goes unrecognized,” the researchers conclude.

The HAVEN system we have developed and validated was able to detect nearly twice as many patients who suffered a cardiac arrest or needed intensive care up to 48 hours in advance, than the next best system.”

Facebooktwitterredditpinterestlinkedinmail