Coronavirus Screening Platform Uses Machine Learning Algorithms to Detect SARS-CoV-2 Structural Fingerprint in Real Time

Botanisol Analytics (Phoenix, AZ, USA) is developing the screening platform that can identify biological threats, including contaminants and pathogens (such as the novel coronavirus), by shining a laser onto the sample. The screener uses next generation Raman spectroscopy, which generates a unique structural ‘fingerprint’ for each molecule or even entire structures, such as whole cells or viruses. The laser interacts with the molecules in the sample, which absorb its energy and emit a higher or lower energy light. A detector collects the new light wavelengths that are emitted and turns them into digital input. A computer interprets the results to produce a Raman spectrum, the unique patterns of scattered light wavelengths represented as peaks. Artificial intelligence can then scan a spectrum to check for any unique patterns related to specific molecules or pathogens.

Using machine learning algorithms, the technology can identify salient differences between healthy and infected cells which could be present in an upper respiratory or saliva sample. As a coronavirus screener, the test output would be if someone is at low or high risk of carrying the virus, allowing people to be quickly grouped for release (low risk) or triage for further testing (high risk), according to BioSpace. The device could screen people for COVID-19 in real time, providing answers about what is in a sample in a matter of minutes. Botanisol’s ‘next generation’ system is also extremely sensitive and is able to detect something in the sample down to parts per million (ppm) or trillion (ppt), thanks to the use of a new type of laser that produces the smallest wavelength able to be transmitted through air under normal conditions, according to the BioSpace report.

“It’s a real-time snapshot of all the molecules in a patient’s sample,” David Talenfeld, JD, MBA, MGM, CEO of Botanisol, told BioSpace. “The device can be calibrated to detect a chemical or pathogen, like coronavirus, whose spectral ‘fingerprint’ is known."

“In the future you could take samples from people getting off of an airplane, quickly screening everyone,” added Talenfeld. “The people identified as ‘high risk’ for carrying coronavirus can be diverted to a special waiting area for further testing.”

Related Links:
Botanisol Analytics