Artificial intelligence (AI) has been used worldwide serving multiple industries, including transportation, finance, education and more. This blog, however, focuses on the role of AI in healthcare, specifically in battling COVID-19 to contain the virus. Since the outbreak, health support call lines have been overwhelmed and new information is being released daily. To mitigate the infection, health authorities and researchers have been evaluating and executing various strategies, one of which is AI technologies.
The first of these highlighted here is chatbot; a computer program used to conduct online chat conversations through voice commands and/or text. Chloe is an example of a chatbot interface, available on both desktop and mobile, created by the Canadian health tech company Dialogue1. Users answer questions for Chloe to identify potential risk factors for COVID-19 tailored to the users, and based on this Chloe provides up-to-date publicly available information and resources1. The system assesses symptoms, provides testing instructions and travel safety advisories (if required), monitors self-isolated individuals to track their disease progress and assists with other concerns relevant to COVID-192. Chloe is fully open-source meaning Canadian health and government organizations can contribute new information to it1.
Contact tracing is not a new solution towards epidemic control as it has been implemented against SARS, measles, typhoid and AIDS, to name a few3. Nevertheless, enhancing it using mobile applications is a recent innovation to curb the spread of COVID-19. AI dependent contact tracing using location tracking has been applied in South Korea, Taiwan, China and Singapore3. In North America, Ron Rivest and his team at the Massachusetts Institute of Technology are developing a system that relies on Bluetooth technology, similar to the Asian countries, to automate contact tracing4. The Bluetooth signals emitted from people’s phones will be recorded on nearby phones as chirps, and if someone tests positive their chirps from the past 14 days will be uploaded onto a database4. If these chirps match the ones recorded on other people’s phones, they will be notified of potential exposure to the virus and what measure they should take4. This form of digital contact tracing will be crucial as the economy reopens and large gatherings commence, making it difficult to control the spread of the virus using only manual contract tracing. Yet, for it to be successful, a large proportion of users have to be willing to download the app to detect the signal and record the chirps.
The third AI technology focuses on supporting society and economy during the reopening phases. A Toronto-based AI firm, Predictmedix, launched mass screening technologies for retailers, restaurants and other businesses to detect symptoms of COVID-195. The infection and protocol screening technology can scan physical spaces to ensure people are maintaining physical distancing and wearing face masks5. If unwarranted behaviour is detected, staff will be notified via an alert on their computer or mobile and they can intercede accordingly5. Interesting to note, the technology is not designed solely for COVID-19, rather it can also screen for other illnesses as well. For instance, the technology can be used to identify cannabis or alcohol impairment and mental illnesses, such as depression and dementia5. I personally see this technology being effective in schools, transits, banks, grocery outlets, factories, gyms and more to enforce physical distancing measures.
These three AI technologies serve numerous purposes. For example, Chloe helps to combat the spread of misinformation due to lack of current and accurate information. As Dr. Daniel Lalla, employee at Dialogue, rightly said, “The more informed the public is, the more likely it is to take preventative actions to stay healthy or consult with health care providers sooner to increase chances of full recovery”1. Contact tracing using AI can speed up the process of tracking potentially infected contacts of positive COVID-19 cases to control the pandemic without the need for large quarantines3. The mass screening provided by Predictmedix could help recover the economy faster, as employees and consumers would be more confident to re-enter the workforce and other places of business5. Hence, automatizing processes not only provides rapid service to the public, but also alleviates the burden on the healthcare system. It allows health professionals to concentrate on activities that require their expertise (optimizing their work), while also improving the healthcare system functioning and capacity2.
While AI holds promising potential for COVID-19 and the future, there are caveats which cannot be ignored. The most evident of these is public privacy since the use of AI for chatbot, for example, may involve “the collection, use, aggregation, analysis and disclosure to third parties of datasets that may or may not include de-identified data”3. Furthermore, using AI for tracking and surveillance, as digital contact tracing and Predictmedix screening technologies propose, raises privacy compliance concerns related to storing of phone location data and monitoring movements via cameras3,5. Therefore, success of the aforementioned and other AI technologies depends on the users trust which is established by respecting their privacy. These tools need to be designed by strictly adhering to privacy regulations for the given geographic regions, prior to adopting them to tackle the COVID-19 pandemic.
References
1. Kolm, Josh (2020, March 12). Tech in Action. Dialogue makes a COVID-19 chatbot. Strategy News. Retrieved from https://strategyonline.ca/2020/03/12/tech-in-action-dialogue-fights-covid-19-misinformation/
2. Canadian Manufacturing (2020, March 12). Artificial Intelligence community to help fight COVID-19. Retrieved from https://www.canadianmanufacturing.com/technology/artificial-intelligence-community-to-help-fight-covid-19-253308/
3. Sookman, Barry (2020, April 14). COVID-19 and privacy: artificial intelligence and contact tracing in combatting the pandemic. McCarthy Tetrault. Retrieved from https://www.mccarthy.ca/en/insights/blogs/techlex/covid-19-and-privacy-artificial-intelligence-and-contact-tracing-combatting-pandemic
4. Foy, Kylie (2020, April 8). Bluetooth signals from your smartphone could automate Covid-19 contact tracing while preserving privacy. MIT News. Retrieved from http://news.mit.edu/2020/bluetooth-covid-19-contact-tracing-0409
5. Patterson, Craig (2020, May 27). AI Company ‘Predictmedix’ Launches COVID-19 Screening Technology for Retailers. Retail Insider. Retrieved from https://www.retail-insider.com/retail-insider/2020/5/ai-company-predictmedix-launches-covid-19-screening-technology-for-retailers