Detection and prediction of spreads of disease outbreak based on syndromic data

Early Warning Systems (EWS) are used to alert health-care providers and authorities about emerging health problems.

Background

Systems based on several different types of input parameters have recently entered the market. With an increasingly mobile world population, the risk of epidemics is rising. Detection of potential disease outbreaks will be useful in preventing and stopping spreads of infectious diseases.

Goal

Develop methods for detection of deviations in spatio-temporal patterns of syndromic data. Develop models for spreads of infectious and non-infectious diseases for the purpose of prediction.

Problems

Suitable statistical methods have improved significantly over the past few decades including such topics as time series analysis of non-stationary data, methods for multi-scale analysis such as wavelets, and methods for analysis of nonlinear data, such as scaling indices. Most of these statistical methods are however either not robust enough or computationally consuming.

New statistical methods that are tailored for the problems at hand will therefore be developed.

Methods

Time series methods, nonparametric smoothing and scale-space methods will be applied. Designing suitable priors for use in Gaussian Markov Random Fields will be an important part of the present project.

Project manager

Vedad Hadziavdic, NST

Project members from partners

Fred Godtlibsen (UiT), Vedad Hadziavdic , Johan Gustav Bellika
(NST)

Researchers

1 Post.doc researcher (NST) and 1 PhD student (UiT)

International collaborators

Collaboration with the data analysis group at the IBM Watson research centre is initiated. It is still not formalized.
Prof. Lasse Holmström, Department of Mathematics and Statistics, University of Oulo, Finland,
Dr. Jörg Polzehl, Weierstrass Institut für Angewandte Mathematik und Statistikk, Berlin, Germany
Prof. Probal Chaudhuri, Indian Statistical Institute, Calcutta, India
Prof. James Stephen Marron, University of North Carolina at Chapel Hill, Chapel Hill, USA.