Researchers use the machine learning algorithm to identify common respiratory pathogens

Reviewed by Emily Henderson, B.Sc.November 11, 2020

The ongoing global pandemic has created an urgent need for rapid tests that can diagnose the presence of the SARS-CoV-2 virus, the pathogen that causes COVID-19, and distinguish it from other respiratory viruses.

Now, Common respiratory from Japan demonstrated a novel system for single-virion identification of common respiratory pathogens using a machine learning algorithm trained on current changes across silicon nanopores. This work can lead to quick and accurate screening tests for diseases like COVID-19 and influenza.

In a study published this month in DHW sensors Scientists at Osaka University have introduced a new system that uses silicon nanopores that are sensitive enough to detect even a single viral particle when coupled to a machine learning algorithm.

In this method, a just 50 nm thick layer of silicon nitride suspended on a silicon wafer added tiny nanopores, which are only 300 nm in diameter. When a voltage difference is applied to the solution on both sides of the wafer, the ions travel through the nanopores in a process called electrophoresis.

The movement of the ions can be monitored by the current they generate and when a viral particle enters a nanopore, it blocks the passage of some ions, causing a transient drop in current. Each dive reflects the physical properties of the particle, such as volume, surface charge and shape, so they can be used to identify the type of virus.

The natural variation in the physical properties of the viral particles had previously hampered the implementation of this approach, however, using machine learning, the team built a trained classification algorithm with known virus signals to determine the identity of new samples.

By combining single-particle nanopore detection with artificial intelligence, we were able to achieve highly accurate identification of multiple viral species. “

Makusu Tsutsui, Studio Senio Author, Osaka University

The computer can discriminate differences in electric current waveforms that cannot be identified by human eyes, which allows highly accurate virus classification. In addition to the coronavirus, the system has been tested with similar pathogens: respiratory syncytial virus, adenovirus, influenza A and influenza B.

The team believes coronaviruses are particularly well suited for this technique as their spiky outer proteins can even allow for separate classification of different strains. “This work will help with the development of a virus test kit that surpasses conventional viral inspection methods,” says latest author Tomoji Kawai.

Compared to other rapid viral tests such as polymerase chain reaction or antibody-based screens, the new method is much faster and requires no expensive reagents, which can lead to better diagnostic tests for emerging viral particles that cause infectious diseases such as COVID-19.