World’s First AI-Powered Diagnostic Test Accurately Identifies Respiratory Viruses in Five Minutes

The ground-breaking virus detection and identification methodology has been described in a paper published in ACS Nano by researchers at University of Oxford (Oxford, UK). The paper demonstrates how machine learning can significantly improve the efficiency, accuracy and time required to identify different types of viruses, as well as differentiate between the strains. The technology combines molecular labeling, computer vision and machine learning to create a universal diagnostic imaging platform that looks directly at a patient sample and identifies which pathogen is present within seconds – similar to facial recognition software, but for germs.

In preliminary studies, the researchers have shown that the test can identify the COVID-19 virus in patient samples and further research determined that the test could be used for diagnosing multiple respiratory infections. In a study to validate the new method that uses AI software to identify viruses, the researchers began by labeling viruses with single-stranded DNA in more than 200 clinical samples. The images of labeled samples were captured using a commercial fluorescence microscope and processed by custom machine-learning software that is trained to recognize specific viruses by analyzing their fluorescence labels, which show up differently for each virus due to their varying surface size, shape and chemistry. The study showed that the technology is capable of rapidly identifying different types and strains of respiratory viruses, including flu and COVID-19, within five minutes and with an accuracy of >97%.

“Our simplified method of diagnostic testing is quicker and more cost-effective, accurate and future proof than any other tests currently available,” said Dr. Nicole Robb from the University of Warwick and Visiting Lecturer at Oxford’s Department of Physics. “If we want to detect a new virus, all we need to do is retrain the software to recognize it, rather than develop a whole new test. Our findings demonstrate the potential for this method to revolutionize viral diagnostics and our ability to control the spread of respiratory illnesses.”

Related Links:
University of Oxford