SEATTLE, Washington — In light of the recent Coronavirus epidemic, the Canadian AI technology BlueDot is getting attention for having identified the outbreak before either the WHO or the CDC. Along with Metabiota, the AI firm notifies public health officials in different countries of its findings using flight records. Significantly, these new AI systems that predict and track infectious diseases may warn vulnerable countries before they break out.
Public Health Surveillance
In past years, the public health surveillance of diseases has been a priority of the World Health Organization (WHO), especially in developing countries affected by common, transmittable diseases, such as Malaria and Diphtheria. Outbreaks were identified by aggregate case studies and reported to public health officials in order to intervene. Identifying investigation gaps and elimination of infectious diseases remain important goals of the international community.
With the help of WHO, several countries have eradicated Malaria and have Malaria-free status, meaning there had not been a reported case within five years. Algeria gained Malaria-free status in 2019, Sri Lanka in 2016 and the Maldives in 2015. Nevertheless, WHO has called for faster discovery of active individual cases, sharing of international data and improved analytical methods to make informed decisions. Enter BlueDot and Metabiota. They are new AI systems that predict and track infectious diseases.
Using news reports and airline tracking data, the Canadian Biosurveillance firm BlueDot correctly identified the Coronavirus outbreak in Wuhan six days before the U.S. Center for Disease Control (CDC) and nine days before WHO. Part of the challenge that WHO and other organizations faced when identifying the outbreak was their reliance on Chinese government reports confirming reported data, which downplayed the severity.
Kamran Khan founded BlueDot in 2014 using an algorithm that analyzes animal disease outbreak reports, airline data and news reports in 65 languages. Once the AI Epidemiologist has generated a report, it is handled by human epidemiologists who analyze the data and make reports. Those reports are then sent to public health officials and hospitals on the front lines of the outbreak. After experiencing the SARS outbreak of 2003, Khan set out to develop an automated system that would prevent such an outbreak from spreading like it did that year.
Using similar flight pattern data, the San Francisco company Metabiota correctly predicted the likely path of the Coronavirus after its outbreak. It eventually made its way to Thailand, South Korea, Japan and Taiwan. Depending on the country’s profile, including the mortality rate, symptoms of the disease and the availability of medical services, the firm also claims to be able to predict the level of “public anxiety” surrounding the disease. As an example, they cited an outbreak of monkeypox in the Democratic Republic of Congo as a “medium” level of public anxiety based on data collected. In contrast, Metabiota rated the threat of Coronavirus in China at a “high” level.
Like BlueDot, Metabiota uses a natural-language process that collects online reports about a potential new disease. The company claims to also be developing a similar tool for social media in order to better track diseases. In theory, this sort of AI technology would use cellphone data to track the movement of people (and the disease) as well as to detect keywords related to the outbreak in common interactions. Using social media would represent a crowdsourcing method of tracking rather than predicting the movement of a virus or disease.
The Future of AI Biosurveillance
Although not eliminating diseases, these new AI systems that predict and track infectious diseases have the potential to save lives by alerting public health authorities of their presence and rerouting resources and attention to slowing their spread. Much like a weather report or hurricane watch can give advance notice of a natural phenomenon, these new AI technologies are being utilized not only by the global community but also by public health officials in developing countries that are most vulnerable to health risks. BlueDot aims, even further, to get the technology into the hands of the individual users so they can stay informed with or without help from health officials.
– Caleb Cummings