Reusable mask worn by a woman in Newcastle

Prototype ‘breathalyser’ test for Covid-19 shows promise

Image credit: REUTERS/Lee Smith

In an ACS Nano paper, a team of researchers have described the development of a prototype device which non-invasively detects Covid-19 in the breath of patients.

At present, collecting a sample for Covid-19 testing involves swabs of the back of the throat and inside of nose with long cotton buds: an experience many people find very uncomfortable. The sample is then analysed for the novel coronavirus RNA via reverse-transcription polymerase chain reaction (RT-PCR).

This is a very time-consuming laboratory procedure which can sometimes be delayed by backlogs by several days, resulting in a Covid-19 positive person transmitting the virus further while waiting for their result. Widespread access to cheap, fast testing could be vital for reducing transmission and mortality rates.

In an effort to meet this need, a team of researchers led by Technion-Israel Institute of Technology’s Professor Hossam Haick have been working on nanomaterial-based sensor which could detect Covid-19 in exhaled breath. The device can be used similarly to a breathalyser.

Previous studies have shown that viruses and the cells they infect emit volatile organic compounds (VOCs) which can be exhaled in breath. Haick and his colleagues made an array of gold nanoparticles linked to molecules which are sensitive to various VOCs; when VOCs interact with the molecules on a nanoparticle, this causes a small change in electrical resistance.

The researchers trained the sensor to detect Covid-19, using machine learning to compare the patterns of electrical resistance signals. In March they obtained data from the breath of 49 confirmed Covid-19 patients, 58 heathy control patients, and 33 patients with lung infections unrelated to Covid-19. The participants blew into the device for 2-3 seconds from a distance of 1-2cm. Where possible, Covid-19 patients also provided samples after they had recovered from the disease.

Once a potential Covid-19 ‘signature’ had been obtained, they tested the accuracy of the device on a subset of participants. The device showed 76 per cent accuracy in distinguishing Covid-19 cases from healthy controls, and 95 per cent accuracy in discriminating Covid-19 cases from lung infections. It was also capable of distinguishing between sick and recovered Covid-19 patients with 88 per cent accuracy.

While further validation studies are needed, even at this stage a device like this could prove useful for quickly screening large populations to determine which people may require further testing. The technique could also be applied to test for other diseases, with the platform trained on the relevant datasets to detect the ‘signatures’ of these diseases.

A hospital in Lyon, France, is engaged in the second trial phase of a similar Covid-19 ‘breathalyser’ which could produce a diagnosis in a matter of seconds.

Sign up to the E&T News e-mail to get great stories like this delivered to your inbox every day.

Recent articles