INFEMAT
- Development of mathematical models for infectious diseases, preventive strategies and vaccination programs
Background
On the basis of the need of modelling infectious diseases and their prevention and the new possibilities for this, a joint research program called INFEMAT was started in 1994 as a cooperation of researchers in three disciplines: infectious diseases (National Public Health Institute, KTL), mathematics (University of Helsinki, Rolf Nevanlinna Institute, RNI) and computer science (Helsinki University of Technology, TKK). The study was funded in part by the Academy of Finland in years 1994-95. Thus far submodels of Haemophilus influenzae type b (Hib) disease process have been developed for the transmission of Hib bacteria in a family setting, including age and antibody responses to carriage and vaccination in young children.
The promising start has made it clear that the basic concept was sound, and that the task should not only be continued but approached with a higher level of input. This was also recommended by the international expert evaluation panel during the evaluation of KTL during 1994-95 organized by the Academy of Finland. The continuation project is part of the From Data to Knowledge (FDK) consortium.
General aim
- develop methods for analysis of Hib infection and the effect of different intervention strategies in different populations, including probability models and simulation of Hib infection and scientific visualization
- develop methods for analysis of Pnc infection and the effect of different intervention strategies, including probability models and simulation of Pnc infection and scientific visualization
- apply and test this approach to other infectious diseases and intervention strategies
- establish a research tradition of infectious diesase modelling in Finland
Aims in the three consortium disciplines
Medicine
- deepen the knowledge on natural course and pathogenesis of Hib and Pnc infection
- test hypotheses of the natural course and pathogenesis of Hib and Pnc infection
- model Hib and Pnc vaccinations
- explore data collection of other vaccine preventable diseases (e.g. measles, mumps, rubella, diphtheria) and sexually transmitted diseases, including infections caused by Chlamydia trachomatis and HIV as well as hospital infections and their need and suitability for modelling
- develop models for evaluation of vaccine and other intervention strategies for infectious diseases in general
Mathematics and statistics
- develop individual level stochastic intensity models for the course of infectious diseases (as examples Hib and Pnc diseases)
- develop spatio-temporal stochastic models for the spread of infectious diseases
- study the use of computer intensive Markov Chain Monte Carlo algorithms in longitudinal Bayesian models
- study the usefulness of Bayesian modelling in complex dynamic systems (e.g. course of an infectious disease) in which the underlying structure is incompletely observed
- combine information from individual and population level models by simulation to study the impact of different vaccine strategies
Computer science
- develop visual programming tools for interactive specification and simulation of the above probability models of infections
- develop methods and tools for visual exploration of both observed and particularly the simulated data
- search for data located in disparate sources
- selection and sorting of data to be visualized
- choice of visualization techniques
- extend from traditional scientific visualization to other medialization techniques, such as sonification
- evaluate the feasibility of genetic algorithms for modeling evolutionary patterns of multiresistant bacteria in hospital environments
- utilize computer networks for distributed processing, e.g. access data from regional registers to run analysis and simulation programs in a supercomputer, controlled by a researcher's desktop workstation
Contact persons
http://www.cs.hut.fi/~tta/research/Infemat.html
Tapio.Takala@hut.fi
4 May 1997