I had an idea while listening to NPR the other day, but I don't know about its viability.
The NPR discussion was about the CDC having reduced capabilities to identify outbreaks of foodborne illness (since doctors have to send it sample data for analysis by experts). Then it occurred to me that maybe a sort of social mobile application might be able to help identify outbreaks.
The app would prompt the user for their symptoms, what they have eaten over the past several days (vegetables, meats, condiments, etc), and their relative location (locations services, postal code, or something) and the data is submitted anonymously.
The the database would have a public API for the data so developers could try to identify correlations and useful information (IIRC much like a game some developers made for identifying DNA mutations).
I don't really have the ambition to put in the research to see whether something like this is viable or not. There are obvious issues with this design since there's no way to identify molecular information until the iBloodTester is invented.
Google has shown that people run flu-related queries when they are affected by flu symptoms so that is a good method for tracking flu outbreaks in time and space.
I'd be a little afraid of volunteer science in this case because I think you might attract a population of hypochondriacs. (At my food co-op it seems 50% of people think they are allergic to wheat and rather than eating rice or beans or skipping the carbs, they buy triple-priced gluten free "breads")
I think something like this would be prone to witch hunts like the guys on Reddit who were looking for boston bombing suspects.
It's an interesting idea, but I wonder if people would expend the effort to accurately log their symptoms/recent behavior without some sort of reward. A paper published earlier this year investigated using machine learning with Twitter data to try to do something similar to what you suggested, though. (the paper: http://www.cs.rochester.edu/~sadilek/publications/Sadilek-Br...).
The NPR discussion was about the CDC having reduced capabilities to identify outbreaks of foodborne illness (since doctors have to send it sample data for analysis by experts). Then it occurred to me that maybe a sort of social mobile application might be able to help identify outbreaks.
The app would prompt the user for their symptoms, what they have eaten over the past several days (vegetables, meats, condiments, etc), and their relative location (locations services, postal code, or something) and the data is submitted anonymously.
The the database would have a public API for the data so developers could try to identify correlations and useful information (IIRC much like a game some developers made for identifying DNA mutations).
I don't really have the ambition to put in the research to see whether something like this is viable or not. There are obvious issues with this design since there's no way to identify molecular information until the iBloodTester is invented.