Laser-Based Breath Analysis Detects COVID-19 Infection

Researchers at the University of Colorado at Boulder and the National Institute of Standards and Technology (NIST) have developed a laser-based breathalyzer technology that can detect molecules in breath samples that indicate the presence of specific diseases, such as COVID-19. The device is called a frequency comb breathalyzer, and using it involves pumping a breath sample into the device where lasers irradiate it at many different frequencies, and mirrors bounce the light around to ensure that the sample is thoroughly investigated. Based on how the molecules in the sample absorb the light, a machine learning algorithm analyzes the millions of data points obtained by the system to provide a yes/no for the presence of molecules that indicate the presence of disease. The technology is non-invasive and very rapid, potentially increasing the speed and simplicity of diagnosing a range of health conditions.

Breath analysis is an intriguing way to determine health, based on the concept that exhaled molecules can reveal underlying health conditions. Such technology has numerous advantages over other diagnostic approaches, including its non-invasive nature, and, in the case of this latest development, rapid analysis. This latest device is called a frequency comb breathalyzer, and uses mid-infrared laser light and AI to perform a spectroscopic analysis of molecules in exhaled breath.

“There is a real, foreseeable future in which you could go to the doctor and have your breath measured along with your height and weight…Or you could blow into a mouthpiece integrated into your phone and get information about your health in real-time,” said Jun Ye, a lead developer of the new device. “The potential is endless.”

The technology was tested during the COVID-19 pandemic in its ability to discern if someone was infected. The researchers collected breath samples from college students who had undergone a PCR test for COVID-19 in the previous 48 hours. Half of the students tested positive for COVID-19 through PCR and the other half tested negative.

They fed the samples into the device and AI analyzed the molecular signature for COVID-19. Strikingly, the breath analysis system matched the PCR results for the students 85% of the time. As PCR is the gold standard test for COVID, these results are encouraging, and the researchers suspect that the accuracy may have been higher if the PCR sample and breath sample were obtained at exactly the same time.

“If you think about dogs, they evolved over thousands of years to smell many different things with remarkable sensitivity,” said Ye. “We are just at the very beginning of training our laser-based nose. The more we teach it, the smarter it will be come.”

Here’s an NIST video about the technology: