The position is part of an interdisciplinary collaboration on writing a system to predict epidemics by detecting deviations from people's regular behavior in large-scale cell phone metadata. The project benefits from a unique data set of anonymized cell phone records for a large population that have been coupled with individual information about onset of influenza-like illness during the A(H1N1) epidemic.
Disease outbreaks, such as Influenza A(H1N1) in 2009, can affect millions of people. Many of those people regularly use digital technology such as mobile phones for communication. These devices create "who-called-whom" records that are collected by the providers for billing purposes, and which can give a unique window into human behavior, for instance by detecting anomalous deviations from people’s regular routines. The post-doctoral fellow would join an interdisciplinary collaboration between experts in computer science, human health and epidemiology to answer the following question:
Can anonymous mobile call records be leveraged to help detect potential epidemics in real-time and follow their trends?The long-term vision of the project is to develop SMART (Surveillance through Mobile Anomalies in Real-Time): an early-warning system through automatic behavioral sensing that could have significant impact on disease surveillance around the world, in particular in the developing world, by helping predict the spread of epidemics and optimize control strategies.
The goals of this post-doctoral fellowship are to build longitudinal models for individual behavior to quantify anomalies in the data, to analyze human contacts over time, to detect and extract useful features in the data that are indicative of disease and to create a prototype system for real-time epidemic detection for a successful model. The candidate will develop expertise in large-scale data analysis, machine learning, systems building and epidemiology.
Scientific environmentThe collaboration actively includes scientists at: Bristol University in the UK (Dr. Leon Danon, epidemiology), the Directorate of Health in Iceland (Dr. Gudrun Sigmundsdottir, epidemiology), and Emory University/Georgia Institute of Technology (Dr. Gari Clifford, machine learning and biomedical informatics). The fellowship is funded by a project grant from the Icelandic Research Center (RANNIS) and the School of Computer Science (SCS), Reykjavik University (RU), Iceland. Emory University also boasts world-class scientists in various areas of health sciences, statistics and computing and other helpful resources during the project.
QualificationsA Ph.D in computer science, statistics, applied mathematics, engineering or a related field is required. Solid programming and data analysis skills as well as interest in life sciences are expected. Experience with machine learning is desired but not required.
The annual salary for the position is $44,000 (negotiable). The maximum term of contract is 30 months.
By January 13, 2016, interested candidates should send their CV, including a list of publications, as a PDF document, a brief statement outlining their suitability for the project and the names, telephone numbers and e-mail addresses of at least two references who are willing to comment on the research potential of the candidate. Please direct applications and inquiries to firstname.lastname@example.org.