COVID-19 response: Digital health and analytics

 

Identifying and detecting potential outbreaks of infectious disease around the world remains our best defence against these threats in Canada. Today, as the world adapts to living with COVID-19, finding the virus and tracking it in our communities is an important way to contain its spread and protect Canadians.

The Global Public Health Intelligence Network System

The Global Public Health Intelligence Network (GPHIN), headquartered at the Public Health Agency of Canada (PHAC), is an early warning system used to identify potential public health threats worldwide, including outbreaks such as avian influenza and SARS (Severe Acute Respiratory Syndrome).

Originally developed by the Government of Canada in collaboration with the World Health Organization (WHO), the GPHIN system was recognized by the WHO as the organization's "single most important reporting source for identifying potential (international infectious disease) outbreaks."

Between 2016 and 2018, PHAC commissioned the National Research Council of Canada (NRC) to build the replacement multilingual text analytics software application for GPHIN. The NRC used its research expertise in natural language processing, artificial intelligence and text analytics to build the new software application.

On December 31, 2019, the GPHIN software application detected the emergence of an unusual respiratory phenomenon in Wuhan, China; this viral outbreak was subsequently termed COVID-19.

The GPHIN software application is meant primarily for PHAC's human analyst team to maintain curated global awareness of emerging public health hazards. PHAC analysts use GPHIN to monitor media sources worldwide and provide organized, relevant information to respond to potential health threats in a timely manner. Other users include non-governmental agencies and organizations, as well as international and foreign government authorities who conduct public health surveillance, including the WHO.

As an open source text analytics platform, the GPHIN software application augments formal public health reporting by harvesting 5,000 to 9,000 articles per day from around the world in 10 languages. Based on PHAC categories, the GPHIN software application applies state-of-the-art machine translation and natural language processing techniques to extract and display information for PHAC analysts to use.

The GPHIN software application is an example of how artificial intelligence tools can be deployed to inform public policy. The NRC monitors the GPHIN software application as part of ongoing technical service but does not play a role in decisions made regarding public health threats.