Follow Datanami:
August 11, 2014

Health Tracker Said to Spot Ebola Outbreak Before WHO

HealthMap, a web-based utility and algorithm used by public health agencies and others to track disease outbreaks and deliver real-time intelligence about emerging infectious diseases, may have spotted the Ebola outbreak in West Africa days before it was officially revealed.

The web site PublicHealthWatch reported on August 10 that the HealthMap algorithm flagged a “mystery hemorrhagic fever” in forested areas of southeastern Guinea nine days before the World Heath Organization (WHO) announced the Ebola epidemic. HealthMap tracks thousands of government and social media sites along with local news reports, then filters irrelevant data, identifies diseases and maps their locations.

“HealthMap goes well beyond the standard mash up and is more like a small-scale implementation of the long-awaited semantic web,” PublicHealthWatch reported. The site creates machine-readable public health information from text indexed by Google News, WHO updates, reports from the Centers for Disease Control (CDC) and other public health agencies.

Created in 2006, HealthMap is run by researchers, epidemiologists and software developers at Boston Children’s Hospital.

The site identifies, locates and tracks outbreaks on a world map and uses color coding to indicate the severity of an outbreak based on public health information identified by the algorithm. The public health data can then be analyzed and visualized by users.

The earliest Ebola cases associated with the current outbreak in West Africa showed up in informal reports in early February. HealthMap is said to have reported on March 14 an outbreak of “mystery hemorrhagic fever” that had already claimed eight victims in Guinea. The site also spotted one of the first local news reports on March 19. That triggered a HealthMap alert.

On March 23, the WHO issued its first public statement about the West African outbreak, which has since spread to Sierra Leone, Liberia and Nigeria, where two deaths and 13 cases have been reported as of August11, according to HeathMap.

Medical personnel in the region began blogging about treating patients with Ebola-like symptoms. Those blogs were mentioned on social media, and HealthMap eventually picked up the trail.

HealthMap researchers describe the site as a “Linux/Apache/MySQL/PHP [programming language] application” that that uses open source tools like GoogleMapAPI for PHP, the Google Translate API and the xajx PHP AJAX library along with filtering software. It disseminates alerts about emerging infectious diseases in nine languages.

Along with WHO, other data sources used by HealthMap include commercial news aggregators, the GeoSentinel travel tracker and food, animal health and even data from the Wildlife Data Integration Network overseen by the University of Wisconsin’s School of Veterinary Medicine.

Observers stress that the HealthMap algorithm is not intended as a replacement for epidemiological analyses by WHO or CDC, but does provide a curated source of data for public health officials.

In the case of the current Ebola outbreak, it also appears to have served as an effective early warning system.

Related items:

Driving MapReduce into the Semantic Web

Big Algorithms to Change the World