The HyperSpectral platform is using AI to fight ‘invisible threats’

In Northern Virginia, a startup is using the world we can’t see to detect threats in our food, health and livelihood.

Alexandria’s HyperSpectral is a spectral intelligence platform that cofounder and CEO Matt Theurer said is hardware agnostic. The technology is designed to eliminate “invisible threats” like pathogens and contaminants by generating insights in spectral data, which includes everything from ultraviolet to infrared light ranges. With this, companies can receive real-time, actionable insights, like if a contaminated food shipment must be stopped, with the help of AI.

With AI, HyperSpectral can analyze hundreds of wavelengths simultaneously so results come back fast, as well as identify very small anomalies.

“The idea of using spectral data is not new, it’s been around for a long, long time,” Theurer told Technical.ly. “But what’s made this practical now is the combination of artificial intelligence, ubiquitous connectivity and the driving need to detect these things at the edge.”

Theurer said that there are many use cases for the technology; but right now, HyperSpectral is focusing on food safety, health and the environment. In food safety, its algorithms can track L: listeria, E. coli, Salmonella and more. In human health, it can track bacterial infections and viruses, like influenza or COVID-19, which requires rapid identification. In the environmental world, it can locate volatile organic compounds or greenhouse gases like methane.

The 10-person company was founded in 2022, and Theurer said that he and the founding team, CTO Paolo Masini and COO Lauren Stack, were inspired by the pandemic. To them, it didn’t make sense to take a test and have to wait a few days for the response, which meant that folks didn’t really know the up-to-date status of their health.

“As we looked at the capabilities of spectral data beyond the visible light range, we realized that there’s a tremendous amount of information there that we could unlock with artificial intelligence and apply across a large number of industries and a large number of use cases,” Theurer said.

The HyperSpectral system, he said, can give results in anywhere from a few minutes to a few hours.

The company is headquartered in Alexandria, though it also boasts an affiliation with the Massachusetts Institute of Technology. Theurer declined to discuss funding or revenue but said HyperSpectral is growing as a team and a company.

The company’s technology primarily runs in the cloud and is based in Python. HyperSpectral uses third-party, commercially available spectrophotometers that meet its requirements for resolution and spectral width. That data is transmitted into the cloud, where it runs through HyperSpectral’s pre-processing and AI algorithms.

Right now, Theurer said HyperSpectral is mostly working with food processors and food safety labs. But it’s also engaging with the federal government via the Biomedical Advanced Research and Development Authority, and he only expects the company to grow. To him, the health of people and the planet is nothing less than existential.

“We feel that a lot of the current methodologies don’t meet the needs to combat the rising occurrence of contamination, of diseases, foodborne illnesses, greenhouse gas emissions,” Theurer said. “Our technology allows us to basically find these contaminants much, much earlier than they are today and at extremely minute qualities before they become a problem.”

This editorial article is a part of Biotech Month of Technical.ly’s editorial calendar.