Powered by AWS, BlueDot Uses Machine Learning to Detect and Respond to Infectious Disease Risks

The world needs a global early warning system for infectious diseases, but governments and private organizations need technology to quickly identify disease outbreaks and mitigate potential risks.

“We’ve been using machine learning to enhance the detection of global threats around the world in near real time,” says Kamran Khan, founder and CEO of BlueDot. BlueDot’s software-as-a-service platform “is gathering information on over 150 different diseases and syndromes in 65 languages and collecting this information every 15 minutes, 24 hours a day,” explains Khan. BlueDot scans official notifications, health forums, and online media for potential threats and cross-references those sources with commercial air-traffic itineraries and other datasets to predict outbreak risk and immediately notify its clients.

BlueDot’s platform has to process a lot of data; that’s where Amazon Web Services (AWS) comes in. “We’ve been using natural language processing and machine learning supported by AWS to extract vital pieces of information—the name of the pathogen, the location and the time of the outbreak, and other contextual data,” says Khan. For example, BlueDot used its platform to detect an outbreak in Wuhan, China—of what would later come to be known as COVID-19. “Using these analytics, we identified the top 20 cities that we thought would be at the greatest risk of impact if COVID-19 were to continue to spread, and spread ultimately outside of mainland China,” says Khan.