Scientists have developed the world’s first diagnostic test powered by artificial intelligence. It can identify known respiratory viruses from a single nose or throat swab in less than five minutes.
New diagnostic tests may replace current methods, such as the lateral flow test for COVID-19, which are limited to testing for only one infection. Otherwise, it is lab-based and either time consuming or rapid and inaccurate.
The novel virus detection and identification methodology is described in a paper published in ACS Nano written by DPhil student Nicolas Shiaelis and Professor Achillefs Kapanidis of the Department of Physics, and Dr Nicole Robb of Warwick University, a visiting lecturer at the University of Oxford. I’m here. of physics.
This paper shows how machine learning can significantly improve the efficiency, accuracy, and time it takes to not only identify different types of viruses, but also distinguish strains.
Nicolas Shiaelis and Dr Robb collaborated with John Radcliffe Hospital to validate the new method. This groundbreaking testing technology combines molecular labeling, computer vision and machine learning to create a universal imaging platform that can directly see patient samples and identify which pathogens are present in a matter of seconds. . It’s a lot like facial recognition software, but germs.
Preliminary studies demonstrated that the test can identify the COVID-19 virus in patient samples, and subsequent studies demonstrated greater than 97% accuracy using the test to detect multiple respiratory infections within 5 minutes. It turned out that it can be diagnosed with
Dr. Robb and Nicolas Shiaelis founded Pictura Bio, a spin-out of the University of Oxford, which is now licensing the technology. They are now seeking further investment to accelerate development and reach the frontiers of healthcare.
“Our aim at Pictura Bio is to turn this method into a diagnostic test by creating a purpose-built imager and disposable cartridge for use in point-of-care testing, with limited input from the user. We are also expanding the number of viruses the model is trained on, and eventually plan to begin investigating other pathogens such as bacteria and fungi in respiratory samples, blood and urine.”
Nicholas Shaelis
Dr Robb said, “The number of cases of respiratory infections this winter has reached a record high, increasing the number of people seeking medical care. Tensions, combined with an aging population, are putting the NHS and its workforce under immense and unsustainable pressure.
“Our simplified method of diagnostic testing is faster, more cost-effective, more accurate and more future-proof than any other test currently available. Rather than developing a test, we simply need to retrain the software to recognize it.Our findings suggest that this method will revolutionize viral diagnosis and our ability to control the spread of respiratory disease. It shows potential.”
Original: Oxford scientists develop test that can identify respiratory viruses in less than 5 minutes
Than: University of Oxford | University of Warwick
