Using the semantic web and social networks to track disease

October 6, 2007

In a recent blog post, Hari writes

Imagine all of us updating a common channel with “de-personalized” information on what afflicts us globally. I can imagine the system to be something like this ..I could “submit” to this service information about what ails me ..and the machine could obfuscate my details , preserving only things like my approximate geographical area and my age and sex and add it to this health information social network.

This brings to my mind a post in which I had written about the CDC and how it has embraced the semantic web. You can see such a concept being extended to the kind of system that Hari is talking about. If there was a system that allowed people to upload information to a database that can capture the right kind of metadata to feed a system like EpiSPIDER, then I don’t see why it would be that difficult to do what Hari proposes. The key lies in making any such submission system simple to use, i.e. any geotagging, semantic information is handled by the system and not the user. Another possibility is to bring healthcare providers into the picture, as then you could actually have a service that pings EMR systems to find out if there are any patterns for infectious diseases, etc.

If our goal here is to just look for patterns and not at specific locations, then people might be more willing to volunteer such information. The challenge will be fighting off those who might game the system or feed it false information, but given that there is nothing commercial to be gained here, perhaps a simple Captcha system would do the trick.

How would social networks play into all this? I am not sure. Companies like Trusera are building social networks around disease information, but that’s targeted at a different goal. Perhaps there could be a way to add geolocation based early warning systems to Twitter, Facebook etc. A CDC app perhaps?

Maybe we should talk to the folks at the CDC