Scientists from Oxford working in the school’s Department of Physics have developed a new type of COVID-19 test, using the SARS-CoV-2 with a high degree of accuracy, directly in samples from patients using a machine learning approach This could help circumvent delivery restrictions, and it also has advantages in detecting actual virus particles instead of antibodies or other signs of the presence of the virus that do not necessarily correlate with an active, communicable case.
The test created by the Oxford researchers also offers significant speed advantages and provides results in less than five minutes without the need for sample preparation. That said, it could be among the technologies that enable mass testing ̵
The technology that makes this possible works by marking virus particles in a sample collected from a patient with short, fluorescent strands of DNA that act as markers. A microscope images the sample and the marked viruses present. The machine learning software then does the algorithmic analysis developed by the team to automatically identify the virus. Differences are used that each individual generates in relation to the fluorescent light emitted due to their different physical surface, make-up, size and individual chemical composition.
This technology, including the sampling equipment, microscopic imager, and fluorescence insertion tools and computational functions, can be miniaturized enough, according to the researchers, to be used almost anywhere – including “businesses, music venues, airports,” and more . The focus now is on starting a spin-out company to commercialize the device in a format that integrates all of the components together.
The researchers expect they can set up the company and begin product development early next year. A device may be approved for use and ready for distribution about six months later. It’s a tight schedule to develop a new diagnostic device, but the schedules have already changed significantly in the face of this pandemic and will continue to do so as we are unlikely to see if they will fade at any time in the near future.