After a relatively quiet summer, there was a worrying increase in the number of corona infections in October last year. The second wave hit our country and demanded new drastic measures.  As a   scientist,  you look for ways to make a contribution based on your own expertise.  Peter Peumans,  CTO Health of imec, also thinks this way.  He  came up  with  a promising new concept to detect the virus in the exhaled air. Here he saw a unique opportunity for imec's chip technology: a silicon sieve, a very tiny sieve, that could capture the virus particles in the exhaled air. That small sieve can then be used to detect the presence of the virus with a very accurate and fast PCR test. The  results of the  first  clinical  studies  were  immediately very promising:  it is  possible to catch and detect the virus in exhaled air.  To ensure that the technology is also workable in practice,  the research group imec-mict-UGent and the imec.EDiT team were asked to study the user experience. 

Now, one year later, we have carried out 30 small- and large-scale experiments in our laboratories in De Krook in Ghent. During these experiments, we combined the expertise of experimental psychologists, product designers and engineers to very quickly build prototypes of the breath test  and then test with real people. MacGyver-wise, we built several test setups in our pop-up lab. The goal was clear:  how can we get people to take the breath test in the most user-friendly and efficient way?  

Figure 1: Pop-up lab at De Krook with (from left to right) Aduén Darriba Frederiks, Jelle Saldien, Wout Duthoo, Klaas Bombeke, Jonas De Bruyne and Nikte Van Landeghem. 

We first wanted to investigate the blowing capacity of the people. But it soon turned out that everyone  was blowing in a random way in our prototypes. In order to ensure that we could better compare the blowing performance of our subjects and to be able to build a more standardized breathalizer test, it quickly became clear that we needed to provide a feedback mechanism. So we put together a dashboard  that  shows when, how hard and how long you have to blow. 

Figure 2: A participant blows into the test set-up with a variable resistance and receives feedback via the dashboard on the screen.

With these first working prototypes, we were able to tackle a number of research questions: what volume and which blowing speed are feasible for the general population? How high can the resistance, introduced by the sieve to catch the breath particles, be exactly? What is the optimal balance between efficiency and comfort?  

In addition, we also looked at ways to motivate and guide specific target groups such as children to perform the breathalyzer test correctly. For this we adapted our software: an animated feedback system on a child's size, with the instruction to blow a pirate ship hard enough so that the ship can reach the treasure chest on the island.  

Figure 3: A child is motivated by an animated story to blow sufficient volume of air into our prototype at the right speed. 

As we gained better control over the blowing protocol for different people, we also noticed differences in the clinical outcomes that we needed to investigate further. For example, the test turned out to detect more virus when people were allowed to 'vocalize'. Using a spectrometer, we were able to confirm that there are indeed many more aerosols (and therefore also virus particles) in the exhaled air when we let our subjects produce certain sounds during the blow-out. This led to a number of new research questions: how can we get people to vocalize while they are still producing sufficient bladder speed? How can we ensure the balance between comfort and efficiency here? How can we ensure that as many particles as possible are blown through the miniscule sieve?  

Other experiments showed that the shape and material properties of the mouthpiece also have a big influence on the number of aerosols that we can measure. Which in turn leads to the question of whether this can have an impact on an even better capture of the virus particles. 

By working very iteratively and interdisciplinary in a team, we succeeded in a short time to solve both fundamental and very hands-on problems and gain new insights. Which ultimately contributed to a new breath test technology that will soon be marketed by miDiagnostics.  

Figure 4: The first Minimal Viable Product (MVP) that is now being further developed for use in the market.  
 
Hopefully, this can contribute to the fight against the SARS-CoV-2 virus in the near future. And it doesn't have to stop there, this approach and technology could also be used to detect other viruses or diseases such as lung cancer early and more easily through the exhaled air.

This project was made possible by the intense collaboration between the various imec groups and with the support of the Flemish Government. 
 
Related articles:  

• https://www.imec-int.com/en/press/imec-signs-licensing-agreement-midiagnostics-commercialize-its-patented-technology-fast-and  
• https://www.imec.be/nl/press/midiagnostics-start-pilootstudie-met-innovatieve-snelle-covid-19-pcr-testoplossingen-op