The scientists first labeled enveloped viruses on throat swabs, by binding short fluorescent DNAs to the surface of virus particles, then surface-immobilized these labeled particles and collected diffraction-limited images containing thousands of labeled particles.
A trained convolutional neural network was used to scan the images to identify the viruses.
The entire process takes less than five minutes.
It’s in “urgent need” since there are various limitations of the current methods of detecting the CCP virus, the paper says.
The “gold standard” reverse transcriptase-polymerase chain reaction (RT-PCR) gives highly accurate results, but takes hours to process. It’s also restricted to specialized laboratories and by supply chain issues.
Isothermal nucleic acid amplification methods can produce results within an hour, but are limited by similar supply chain issues as RT-PCR.
Rapid immunoassay-based antigen-detecting tests exist for some viruses but generally have low sensitivities.
“Unlike other technologies that detect a delayed antibody response or that require expensive, tedious, and time-consuming sample preparation, our method quickly detects intact virus particles, meaning the assay is simple, extremely rapid, and cost-effective,” said Achilles Kapanidis, professor of physics at Oxford.
Nicolas Shiaelis, a doctoral student at the University of Oxford involved in the research, said the extremely rapid test “can make mass testing a reality, providing a proactive means to control viral outbreaks.”