Researchers from the University of Oxford announced in February 2023 that they had developed a new diagnostic test to replace current methods for virus detection. The paper, published in ACS Nano, explores how machine learning can improve diagnosis of not only different types of viruses, but also different strains. This approach could be used instead of lab-based and time-consuming testing capabilities.
In collaboration with the John Radcliffe Hospital, researchers combined “molecular labelling, computer vision, and machine learning” to develop a universal diagnostic imaging platform that identifies a pathogen in seconds: “much like facial recognition software, but for germs”. Initial results show that the test can identify the COVID-19 virus in patient samples, with further research showing that it could diagnose “multiple respiratory infections”.
The research began with 200 clinical samples from the hospital. These were labelled and then captured by a fluorescence microscope and processed by a machine-learning software trained to recognise viruses by analysing their fluorescence labels. With an accuracy rate of above 97%, the technology promises to offer a rapid alternative to current standards.
Dr Nicole Robb and DPhil student Nicolas Shiaelis founded a University of Oxford spinout called Pictura Bio, which now licences the technology. They now seek further investment to accelerate development and deployment. Shiaeles stated that the aim is to “turn the method into a diagnostic test by creating a dedicated imager and single-use cartridge for use in point-of-care testing, with limited input from the user”.
“We are also expanding the number of viruses that the models are trained on and will eventually start looking at other pathogens.”
A current need
Dr Robb emphasised that the “number of people seeking medical help” has increased due to “record-breaking highs” of respiratory infections. Combining this with “the COVID-19 backlog” and NHS pressures, the service is feeling “immense and unsustainable pressure”.
“Our simplified method of diagnostic testing is quicker and more cost-effective, accurate, and future proof than any other tests currently available”.
Dr Robb suggests that “all we need to do” to detect a new virus is to “retrain the software”, rather than develop a new test. These advances in diagnostic capabilities are intended to alleviate healthcare pressures from a collection of viral threats. How can vaccines contribute to these efforts? For more on vaccine technology and novel viral response developments come to the World Vaccine Congress in Washington this April.