Artificial intelligence (AI) has come to bear on many of our responses to SARS-CoV-2 (COVID-19), and will continue to help us tackle future pandemics. A recent case saw AI making a contribution to face masks’ comfortability, monitoring particulate matter in the air through a smart connected sensor in the masks. This research was published in ACS Nano in 2021.
Image Credit: ImageFlow/Shutterstock.com
During COVID-19, many people have gotten used to wearing face masks to protect their communities and themselves. But face masks are often slightly uncomfortable, particularly during exercise.
To tackle this problem, researchers have developed a dynamic respirator that uses AI to help it modulate pore sizes in response to real-world conditions as they are changing. Personal and environmental factors are accounted for, with the mask able to monitor the user’s level of exercise as well as air pollution levels.
This means that users can breathe easier when pollution and other particulate matter are less concentrated in the air around them, as the mask slightly opens its pores to allow more air to pass through.
While the whole world became used to face masks as COVID-19 spread to nearly every country in 2020, many people have been wearing them for decades for several reasons. People with respiratory problems use face masks to filter out harmful pollutants, and they have been used to prevent the spread of flu and other viruses in many parts of the world for years.
Sometimes, especially for people filtering out air pollution, high levels of filtration are not always needed. There are also times when the risk of viral infection is lower, such as when people are alone outdoors or if there is a low rate of infection.
However, masks generally cannot change under changing conditions. Instead, breath is exhaled and then trapped under the mask, creating mild discomfort for wearers.
To tackle this, researchers created an air filter whose micropores expand to allow air to pass through when an air pump and microcontroller chip stretch the filter. Wireless connectivity links the device to a cloud-based AI software that reacts to the levels of particulate matter present in the air and changes in the user’s patterns of breath during exercise.
The AI software enabled the mask to respond to users’ individual breathing patterns, and could be used to develop more personalized face masks in the future, said the team.
AI for Future Pandemic Responses
A recent review by Harvard Medical School and Harvard Business School researchers in the United States sought to identify the key areas that AI could support in pandemic preparedness and response.
The review’s authors identified six main use cases for AI in our fight against future pandemics:
- Forecasting dynamics of infectious diseases and the effects of possible interventions
- Surveillance and detecting outbreaks
- Monitoring public adherence to health recommendations in real time
- Detecting flu-like illnesses in real time
- Triage and diagnosis in a speedy manner
- Prognosis of illnesses and patients’ responses to treatment
The Organization for Economic Cooperation and Development (OECD) has also published a summary of AI’s contributions toward our fight against COVID-19.
The OECD also listed multiple distinct areas in which AI had been effectively enlisted to help tackle COVID-19:
- Learning about the virus and speeding up medical research on treatments
- Detecting the virus and diagnosing patients, and predicting how it would evolve
- Helping to slow the virus’s spread with surveillance and contact tracing
- Personalizing public health information and helping people to understand the virus
- Monitoring society’s slow recovery from the virus and improving early warning methods for the future
The OECD also stated that policymakers should ensure that medical, molecular, and scientific datasets and models could be shared on collaborative platforms so that AI researchers would be able to build tools for the medical community. Data is the lynchpin of an effective AI-based solution, as more data essentially makes an AI more intelligent.
At the same time, said the OECD, policymakers also have a role to play in ensuring medical researchers have access to advanced computing capabilities such as AI and machine learning (ML).
Ensuring that AI systems were only developed in line with best practice guidelines with regards to respect for human rights, privacy, transparency, and security is also strongly recommended by nearly every body and researcher.
AI solutions should be explainable and robust, and the people involved in developing and using them should remain accountable. In the case of AI solutions developed and used with public money, these requirements are even more vital.
Will AI Prevent the Next Pandemic?
AI is already rapidly accelerating research on viruses, as it is able to process and analyze huge amounts of data relatively quickly.
AI tools for text and data mining can collect information about a virus’s history, how it transmits, and then suggest diagnostics and management measures incorporating lessons learned from previous pandemics.
Continue reading: Could Service Robots be the Future of Social Distanced Hospitality?
References and Further Reading
OECD (2020) Using artificial intelligence to help combat COVID-19. [online] Available at: https://www.oecd.org/coronavirus/policy-responses/using-artificial-intelligence-to-help-combat-covid-19-ae4c5c21/.
Shin, J. et al. (2021). Dynamic Pore Modulation of Stretchable Electrospun Nanofiber Filter for Adaptive Machine Learned Respiratory Protection. ACS Nano. Available at: https://doi.org/10.1021/acsnano.1c06204.
Syrowatka, A. et al. (2021). Leveraging artificial intelligence for pandemic preparedness and response: a scoping review to identify key use cases. npj Digital Medicine. Available at: https://doi.org/10.1038/s41746-021-00459-8.