[Using big data for disease surveillance and drug safety monitoring]

Using big data for disease surveillance and drug safety monitoring. The ongoing global growth in internet usage and data sharing has generated tremendous amounts of "big data" that can be analyzed for public health purposes. In addition to electronic medical records, new paradigms are emerging, leveraging the combination of patient-generated data with data from traditional health systems. As the two data sources have complementary strengths - high veracity in the data from traditional sources, high velocity and variety in the data from patient-generated data - they can be combined to build more robust public health systems. Here, we will focus on two areas of public health: infectious disease surveillance, and adverse drug event monitoring. The problem with these data sources are that they are completely unstructured and highly context-dependent, posing essentially a machine learning challenge. Some of the recent examples in these two domains indicate that the technical challenges can be solved, but as long as the two systems are separate, they will be constrained by their intrinsic limits.