The immense connectivity that social media offer is perhaps the most attractive aspect for users, while the “Big Data” that social media offer is perhaps the most attractive aspect for advertisers. Now, both aspects are being used for something very different: tracking disease.
User-generated content is what drives the most popular social media platforms like Facebook, Twitter, Instagram, and Pinterest. Both user profiles and the content that those users share contain “Big Data” like the user’s age, preferences, interests, locations, friends, emotional state, financial state, online usage, favorite websites, and educational level- just to name a few.
Advertisers were some of the first to realize the potential of the personal information that users voluntarily divulge in order to access certain sites, and have traditionally used such information to tailor and target ads to specific audiences.
Beyond online advertising, the exact same connectivity infrastructure that drives social media also has the potential to track contagious diseases like Ebola and polio viral infections, as well as hereditary conditions like heart disease.
Moreover, the same infrastructure can be used model the spread of an epidemic, and even alert populations at risk.
Redefining “Gone Viral”
CrowdBreaks is one example of a disease surveillance system that utilizes public data from Twitter to acquire information on diseases based on what people are tweeting.
Tweets are filtered using a machine learning algorithm that identifies the tweets relevant to a specific disease. Because these algorithms can be trained to use information to make predictions, such techniques have enormous potential to identify populations at risk based on Big Data and provide early warning and perhaps even treatment to those demographics.
Aside from social media, location data from cell phones is especially important in tracking and modeling epidemics in remote regions of developing countries. Because medical resources and equipment are often limited or unavailable, taking advantage of existing infrastructure that populations already use is crucial in combatting the spread of diseases in remote areas. Teams are able to use cell phone calls to pinpoint the location of possibly infected individuals and based on their location relative to the nearest cell phone tower, model their movement to project infection rates and alert populations in the vicinity.
Another example is Qlik, a Big Data analytics company that created an app to track the 2014 Ebola outbreak. The platform relied on data from ports and other transportation hubs in conjunction with cell phone data to project infection rates.
For most of us, it’s normal to post a photo on Instagram of a loved one recovering in the hospital, or post a status update on Facebook wishing a friend “get well soon”. In the future, will machine learning algorithms like the one that CrowdBreaks utilizes collect and analyze such posts to offer targeted healthcare?
Instead of targeted shopping ads, can you imagine a future where pop-ups read “Based on your recent activity, it looks like heart disease runs in your family”? The same pop-ups could potentially use your location data to provide information on treatment centers, physicians, and pharmacies, or even suggest vaccines or medication.
Or, during Flu Season, we may receive a text message alerting us to the presence of infected individuals nearby, and warn us not visit certain locations.
While there are ethical concerns surrounding how the personal information that we share via social media and cell phone calls is used by third parties like advertisers, the same information and infrastructure can be used to improve public health campaigns and to fight the spread of infectious diseases. Information, like technology, is a tool, and as technology makes us increasingly connected, social media and Big Data have the potential to help information on preventative healthcare spread as quickly as disease.