Related Experiment Video
Updated: Aug 16, 2025
04:17
DNA Virus Detection System Based on RPA-CRISPR/Cas12a-SPM and Deep Learning
Published on: May 10, 2024
842
Virus Detection and Identification in Minutes Using Single-Particle Imaging and Deep Learning
Nicolas Shiaelis1, Alexander Tometzki1, Leon Peto2,3
1Biological Physics Research Group, Clarendon Laboratory, Department of Physics, University of Oxford, OxfordOX1 3PU, United Kingdom.
ACS Nano
|December 21, 2022
Summary
A new deep learning method rapidly identifies viruses from microscopy images in under 5 minutes. This convolutional neural network approach offers a fast and sensitive alternative for diagnosing viral infections, including COVID-19 variants.
Area of Science:
- Virology
- Biotechnology
- Artificial Intelligence
Background:
- Viral outbreaks like COVID-19 necessitate rapid and sensitive diagnostic tools.
- Traditional methods often involve lengthy procedures like purification and amplification.
Purpose of the Study:
- To develop a rapid virus detection and identification methodology.
- To utilize deep learning for distinguishing between different viruses from microscopy images.
Main Methods:
- Employing a convolutional neural network (CNN) to analyze microscopy images of fluorescently labeled intact virus particles.
- Achieving labeling, imaging, and identification in under 5 minutes without sample lysis, purification, or amplification.
Main Results:
- The CNN accurately differentiated SARS-CoV-2 from negative samples and other respiratory pathogens (influenza, human coronaviruses).
- The method successfully distinguished between closely related influenza strains and SARS-CoV-2 variants.
- The system allows for easy incorporation of new pathogens via software updates.
Conclusions:
- Single-particle imaging combined with deep learning presents a promising, rapid alternative to conventional viral diagnostics.
- This technology has the potential for significant impact in managing future infectious disease outbreaks and pandemics.
