Extracting More Information from Exhaled Breath

Exhaled breath can provide a treasure trove of health information, offering a noninvasive window to both respiratory microenvironments and systemic physiological states. But collecting such data is a challenge.

In 2024, Caltech's Wei Gao, professor of medical engineering, and members of his laboratory developed a prototype for a smart mask that uses clues from exhaled breath to monitor for a range of medical conditions, including respiratory ailments such as asthma and post-COVID-19 infections. Now, he and a team of researchers have updated the mask to make it more sustainable, to collect information for longer periods of time, and to evaluate additional health conditions and changes.

"We achieved long-term extension of usability through advances in materials science and system-level engineering design," says Wenzheng Heng (PhD '25), a postdoctoral scholar research associate in medical engineering and lead author of a paper published March 16 in Nature Sensors that outlines the team's recent work. "Building on the original mask, these upgrades enable storage and handling practicality, enhance sensing stability, and achieve energy autonomy."

The first thing a smart mask does to collect data is cool exhaled breath using a hydrogel—a water-based substance that can retain water while maintaining its structure—and capture the resulting liquid, called exhaled breath condensate (EBC). Once EBC is collected, sensors analyze the samples for biomarkers of health, such as metabolites, pathogens, and inflammatory indicators, and transmit the data wirelessly to a personal phone, tablet, or computer. But in previous versions of the mask, the hydrogel dried up after a few hours, making continuous monitoring over time difficult.

"The smart mask is a low-cost way of providing continuous health care monitoring," Gao says, noting that the cost of materials for the new platform, called EBClite, is roughly $1 per mask. "We want people to be able to wear this system for longer periods of time, like over days or a week."

With researchers from Caltech, Johannes Kepler University in Austria, and The Hong Kong University of Science and Technology, Heng and Gao (who is also a Heritage Medical Research Institute Investigator) developed a new lithium chloride-infused hydrogel that is anti-drying and also easily rehydrated, maintaining functionality over many days.

Next, the team looked for ways to make the sensors—which can analyze the EBC for chemicals in one's breath in real time—operate longer with improved stability in the high-humidity environment of the masks.

"The sensing process is in a very humid environment, so, sensor-wise, it's also very critical to get the materials right," Gao explains. The team encapsulated the sensors in a flexible multilayer material to improve its ability to withstand the many cycles of repeated dry–wet conditions that occur during prolonged use of the mask.

Finally, the researchers wanted to eliminate the use of the battery found in the previous model. The mask is now powered solely by an ultrathin, integrated solar cell, enabling long-term operation without the need to recharge or replace a battery.

"We can realize battery-free continuous operation during all different types of indoor activities," Gao says. "The solar cell has high-efficiency energy harvesting even in weak light; you don't have to stand under very strong California sunlight for it to work. The entire mask is now very durable and long lasting."

Once the new smart mask was built, the team aimed to extend its use beyond disease assessment by monitoring the chemical lactate in healthy volunteers. Lactate can be informative for many health conditions and is also a marker of athletic performance, as it's a primary fuel for cells during exercise where it can be used to assess tissue oxygenation and metabolic stress.

"We showed that when the participants do different physical activities, we see the lactate increase in the exhaled breath, very similar to what we see in blood tests," Gao says. "Two years ago, we tested mostly on airway inflammation, but we want to make a daily-wear mask so they can benefit anyone who wants to monitor their health data easily and noninvasively."

The team also monitored for lactate after study participants consumed carbohydrates; monitoring of the two can assess energy metabolism.

"What excites me most is that this platform enables noninvasive and repeatable monitoring of EBC lactate that provides insight into blood lactate dynamics, supporting longitudinal tracking of lactate variability and clearance under exercise- and diet-induced metabolic changes, with potential utility in clinical settings," says Heng, whose doctoral thesis work was done on the smart mask in the Gao lab.

Compared with other biofluids, including sweat and saliva, EBC collection is entirely passive, making it uniquely suitable for user-friendly health monitoring across many populations, including children, elderly individuals, and critically ill patients. Gao and his team have begun early deployment efforts in low-resource settings, working with the Gates Foundation to explore distribution of the masks in parts of Africa.

"We are now developing a simplified version of the mask to take to Africa that can monitor for tuberculosis, which is a very interesting and meaningful application," Gao says. Tuberculosis (TB), a contagious lung infection caused by bacteria, is a critical public health crisis on the continent, with a third of all TB deaths occurring there.

With the ultimate goal of making the smart mask available to the public, Gao says they will continue to investigate other clinically relevant biomarkers that the sensors could monitor for and analyze.

"We really hope make this platform more general so it can be applied to many different health conditions in our daily life," he says.

The Nature Sensors study titled "A battery-free smart mask for long-term exhaled breath biochemical sensing" was funded by the Heritage Medical Research Institute and the Horizon 2020 research and innovation program of the European Union. Other Caltech authors include graduate students Canran Wang (MS '23) and Yonglin Chen, and postdocs Moon-Ju Kim and Gwangmook Kim, in addition to Wenying Tang, a former visiting graduate student, and Jihong Min (PhD '24), a former postdoc at Caltech.