what's that smell?

This electronic body part can identify those most vulnerable to Covid-19

By detecting chemical biomarkers in people's breath, this sensor can sniff out what lung diseases a person might have.


Human bodies give out all kinds of secret signals that our senses can't detect, like chemical biomarkers in breath called breathprints. Scientists believe that breathprints could diagnosis different Covid-19 complicating lung diseases like lung cancer, asthma, or COPD. Detecting these biomarkers is more important now than ever and scientists have designed a flexible and cheap sensor that helps them do just that.

Using plastic and carbon nanotubes, researchers have designed a sensor called an electronic nose, or e-nose for short, capable of sniffing out chemical compounds associated with different lung diseases and providing an analysis of the compound in just a few minutes with a high level of accuracy. Compared to previous e-nose designs, this proof-of-concept model is compact, reusable, and capable of operating at room temperature.

As patients with pre-existing lung conditions, especially those with COPD and asthma, are particularly susceptible to complications associated with Covid-19, a simple model like this could be used to quickly identify and protect those most vulnerable during this time.

In a study published earlier this month in the journal Advanced Healthcare Materials, a European team of material scientists writes that part of what makes their sensor unique and easier to reproduce is their use of a thin, flexible and strong form of carbon called a carbon nanotube. As its name suggests, these pure carbon structures are in the shape of small tubes and can form together in ropes to create incredibly light and strong structures. It's these carbon nanotubes, along with thin layers of plastic, that the researchers used to build the structure of their sensor.

"CNTs [carbon nanotubes] were synthesized by aerosol chemical vapor deposition and deposited in the form of thin transparent and conductive films. This technology is highly reproducible, easily scalable, and allows applying films of nanotubes to any surface," Albert Nasibulin, a co-author on the study professor at the Skolkovo Institute of Science and Technology, said in a statement.

The authors write that while carbon nanotubes have greater stability than typical e-nose materials and have been used previously in studies to build gaseous sensors, they had yet to be widely used in the construction of e-noses.

While it doesn't look much like a real nose, the eight electronic sensors on this e-nose are capable of sniffing out chemical biomarkers of lung diseases in people's breath.Sonia Freddi et al / Advanced Healthcare Materials

With their e-nose constructed, the researchers then set out to test it on real, human breath. The team enlisted a group of twenty-one participants (twelve with previously diagnosed COPD and nine healthy subjects) and had them breathe through straws into plastic bags with the sensors inside. These bags were then sealed to prevent cross-contamination while the sensor took three minutes to identify the samples. However, full recovery of this information took up to 15 hours.

The researchers write that the sensors were able to detect higher levels of nitrogen dioxide (a biomarker for COPD) in all COPD patients in the trial, allowing them to correctly differentiate between COPD and healthy patients.

"The sensor is [so] accurate and sensitive that we can even differentiate between different people," coauthor and senior researcher at the National Research University of Electronic Technology in Moscow, Ivan Bobrinetskiy, tells Inverse. "This opens the way for personal encoding of information on breath (in addition to finger trace and eye iris)."

While the success of this trial is an important stepping stone for future carbon nanotube sensors being implemented and more quickly produced, the results of this study remain just a proof-of-concept among a small trial group. The researchers write that larger clinical trials will be necessary before this design is ready to be used on the frontlines.

"For some gases it can detect the ppb level (one particle per billion)," says Bobrinetskiy. "If the pattern of breath from people with COVID-19 would be known it can be used... [But] in [the] paper we talk only about chronic disease that takes time to develop in a person. In [the] case of Covid it develops very fast and to the time the breath will be changed it may be too late."

Abstract: A sensor array based on heterojunctions between semiconducting organic layers and single walled carbon nanotube (SWCNT) films is produced to explore applications in breathomics, the molecular analysis of exhaled breath. The array is exposed to gas/volatiles relevant to specific diseases (ammonia, ethanol, acetone, 2-propanol, sodium hypochlorite, benzene, hydrogen sulfide, and nitrogen dioxide). Then, to evaluate its capability to operate with real relevant biological samples the array is exposed to human breath exhaled from healthy subjects. Finally, to provide a proof of concept of its diagnostic potential, the array is exposed to exhaled breath samples collected from subjects with chronic obstructive pulmonary disease (COPD), an airway chronic inflammatory disease not yet investigated with CNT-based sensor arrays, and breathprints are compared with those obtained from of healthy subjects. Principal component analysis shows that the sensor array is able to detect various target gas/volatiles with a clear fingerprint on a 2D subspace, is suitable for breath profiling in exhaled human breath, and is able to distinguish subjects with COPD from healthy subjects based on their breathprints. This classification ability is further improved by selecting the most responsive sensors to nitrogen dioxide, a potential biomarker of COPD.