AI camera tech could help quickly identify serious infections
It might soon be possible to measure a patient’s pulse, breathing and blood pressure simply by scanning their face, according to a new thesis from the University of Gothenburg.
Those who visit healthcare centres in connection with sore throats, coughs and fever make up a large group of patients. In many cases, these infections are self-healing and harmless, but there are also serious and potentially life-threatening conditions that can start with similar symptoms, such as pneumonia, COVID-19 and Lemierre syndrome. It is therefore important to assess the severity quickly and offer effective treatment if required, but not prescribe antibiotics unnecessarily.
“Finding the needle in the haystack is important when it comes to infectious diseases,” said Stefan Malmberg, the author of the new thesis. “Most people don’t benefit from antibiotics, but in the case of serious infections, timely and appropriate treatment can save lives.”
The severity of an infection is assessed on the basis of vital signs — in other words, the main signs of an individual’s physical condition — which are currently measured using several different instruments. However, Malmberg’s thesis describes a combination of camera technology, software and AI that has the potential to produce equivalent results by scanning the patient’s face for 30 seconds.
The thesis documents how the camera technology was clinically tested during the pandemic on more than 200 patients with suspected COVID-19. The technology provided data on patients’ heart rate, oxygen saturation, respiratory rate and blood pressure, and was shown to improve both severity assessment and diagnosis.
While the results need further validation, including in terms of measurement accuracy, they have been described as promising and could lead to a future tool for quickly assessing the severity of acute infection and other conditions.
“The new AI method means that measurements are faster, more convenient for the patient, easier for the healthcare provider and involve less risk of infections being spread via measuring equipment,” Malmberg said. “This type of research is crucial for the development of new healthcare technologies.”
Machine learning identifies 800,000+ antimicrobial peptides
An international research team has used machine learning to search for antibiotics in a vast...
AI platform makes microscopy image analysis more accessible
DL4MicEverywhere makes artificial intelligence (AI) accessible for analysing microscopy images,...
Potential pandemic prevention strategy raises ethical dilemmas
Locking down the most at-risk group of people, while simultaneously promoting infection in other...