How AI Can Assist in Rapid COVID-19 Testing

Research from UC Davis and Allegiant Travel Company has discovered a new rapid testing method for COVID-19. The test, which pairs artificial intelligence (AI) and mass spectrometry, could provide faster, more accurate results. This accuracy and speed could be a helpful tool in the ongoing fight against COVID-19.


Rapid testing is essential for effective anti-COVID responses. These tests must be fast enough to enable quick responses and prevent outbreaks, and be accurate enough to catch all potential carriers without false positives. That’s proved a challenge throughout much of the pandemic, but this new solution could provide a way forward.


Heightened Test Accuracy


This new COVID test starts by testing nasal swabs with a tool called a mass spectrometer. Mass spectrometers measure the mass-to-charge ratio of a sample, which can help identify different compounds and particles. Once the spectrometer has identified these compounds, a machine learning algorithm analyzes the results.


The AI platform, called Machine Intelligence Learning Optimizer (MILO), looks at spectrometry peaks and signals to identify patterns. From these patterns, MILO has learned which signals correspond to the presence of SARS-CoV-2, the virus that causes COVID-19. It can then tell if a patient has COVID with remarkable accuracy.


According to the project’s research paper, the test was 100% accurate in identifying positive cases and 96% accurate with negative samples. That gives it an overall accuracy of 98.3%, making it as good as or better than the tests health care workers currently use.


Faster Testing


This new test also works faster than the available and similarly accurate alternatives. The entire process takes roughly 20 minutes, mostly due to the speed of the MILO platform. MILO can analyze spectrometry results in a fraction of the time than manual processes can, enabling far more timely and effective responses.


This isn’t the first time that AI has improved the efficiency of medical processes. A 2019 study found that AI can detect breast cancer with comparable accuracy to radiologists in much less time. Machine learning algorithms are far better than humans at making connections between data points and recognizing patterns, making them faster in these applications.


With tests this rapid, businesses requiring negative tests for employees or guests could enforce these policies with minimal disruption. Airlines, conferences, and other areas where crowds gather could ensure they don’t facilitate outbreaks and do so without long waits. Pandemic responses would then be far more effective.


Flexibility for Future Applications


Another benefit of this new testing system is that it can adapt to detect other pathogens as well. The combination of mass spectrometry and AI makes it relatively easy to adjust the technology to look for other signals. As new virus variants emerge, this flexibility could be indispensable.


For example, while current vaccines seem to be effective against the novel delta variant, tests are less promising. The variant carries unique biological markers that many current testing methods may miss. Since this new system can quickly tailor itself for different signifiers, it could maintain its accuracy with emerging variants.


This flexibility will continue to be helpful with other diseases. Researchers could use it to test for other contagious diseases, even new ones, to prevent future outbreaks. Widespread application of these tests could then help stop future pandemics.


Technology Can Help End the Pandemic


New technologies alone will not bring about the end of the pandemic. Still, breakthroughs like this give medical authorities a better chance at enacting timely, effective responses to emerging trends. The world could then start to move past COVID-19.


Medical technology is advancing at a record pace. As more organizations embrace tools like AI, they’ll find more solutions to more problems, equipping the world for future challenges. How long the pandemic will last is still uncertain, but these technologies paint a promising picture of the future.

Anindya Chowdury
Anindya Chowdury
MERN-Stack Web Developer trying to C Rust. Also writing articles sometimes.

Latest articles

Related articles

Leave a reply

Please enter your comment!
Please enter your name here

This site uses Akismet to reduce spam. Learn how your comment data is processed.