Abstract

The majority of Internet users turn to the web for information when they have a medical concern. The data generated while users seek such information, and more generally when browse the Internet for work and pleasure, represent a potential boon for medical research. During the past decade these data have proven valuable where the most patient activity happens online, where internet data provides a more sensitive indicator than that attainable from traditional sources, and where reports from people suffer from significant reporting bias. However, extracting relevant insights entails overcoming several challenges, including processing of unstructured text, identification of relevant cohorts, causal inference from retrospective analysis, and the preservation of privacy. In my talk I will exemplify some of the ways we have developed for overcoming the challenges in Internet data and the insights we have gained from these data. I will show how people who share the same medical condition can be identified from their anonymized search engine queries, making possible early screening for ovarian and cervical cancers as well as discovery of risk factors for disease. I will discuss the use of Internet data in evaluating the success of influenza vaccination campaigns, and for tracking the parameters of seasonal epidemics.