Hardly a day goes by that artificial intelligence (AI) is not mentioned in the popular press. In our industry, AI is found in myriad applications, from predictive maintenance – anticipating failures before they actually occur – to man-to-machine (M2M) systems that optimize building comfort while minimizing energy consumption.
Now that the relationship between the transmission of coronaviruses, including SARS-CoV-2, and filter efficiency/ventilation rates is becoming better understood, smart HVAC controls will undoubtedly help reduce infections.
However, the potential for AI to contribute to controlling the COVID-19 pandemic goes well beyond HVAC controls since it is now at the forefront of vaccine development.
Although Edward Jenner is generally considered the “father” of modern-day vaccination, demonstrating in 1796 that cowpox (like smallpox, caused by the vaccinia virus, hence the term vaccine) inoculation conferred immunity to smallpox. Even then, it took two years before an actual smallpox vaccine was developed. In the intervening two centuries, one of the underlying characteristics of vaccine development has been the prolonged length of time required to produce a safe and effective new vaccine.
According to the World Health Organization (WHO), as of October 19, globally there were 44 vaccines for COVID-19 in development, 10 of them in Phase 3 clinical trials. Considering the virus was unknown less than a year ago, that is an amazing accomplishment that would not have been possible without AI.
Specifically, there are at least eight different types of vaccines being investigated, including inactivated virus, DNA, RNA, and other proteins. So far, AI – machine learning systems and advanced computational modeling analyses – has helped the researchers to predict responses to different agents, and to track the voluminous amounts of data produced by these efforts. AI can, for example, rapidly sort through thousands of protein characteristics, target sites, etc., and use sophisticated algorithms to predict immune responses.
Of course, AI systems are not just limited to vaccine research in the fight against COVID-19. They are also being used to improve clinical decision-making, and a tool that predicts acute respiratory distress syndrome, one of the possible complications from COVID-19, as well as other diseases, was recently developed by a team led by Prof. Suchi Saria, director of the machine learning and healthcare lab at Johns Hopkins University. According to Prof. Saria, this tool will soon be available to hospitals nationwide.
Of course, AI cannot replace human trials. It can, however, rapidly analyze the results of those trials, so that successful candidate vaccines can be distributed more quickly.
##########
A regular contributor to HPAC Engineering and a member of its editorial advisory board since 2012, Larry Clark, LEED AP, O+M, is a principal at Sustainable Performance Solutions LLC, a south Florida-based engineering firm focusing on energy and sustainability. Email him at [email protected].