De Almeida Simones, JM;
(2007)
An agent-based approach to spatial epidemics through GIS.
Doctoral thesis , UCL (University College London).
Preview |
Text
DeAlmeidaSimoes_thesis.Redacted.pdf Download (63MB) | Preview |
Abstract
This study introduces a bottom up approach to model the complex phenomena involving the spread of epidemics in a human population with spatial dependence. To pursue this, several concepts and technologies are assembled together such as Artificial Life, Complex Networks, and Geographical Information Systems (GIS). At the beginning of this project, a Cellular Automata (CA) model was developed, but that was replaced later by an agent-based model (ABM) as it seemed more ad equate for this particular context the main target was relaxing the assumptions of homogeneous space and of random mixing of individuals. The model is split into a movement model and an infection model. The movement model uses network concepts and combines different ranges of movements to emulate the movement of individuals in human populations. The infection model is based on a classic SEIR ( Susceptible- Latent-Infectious-Removed) model and it is flexible enough to accommodate different infectious diseases. Both these models are probabilistic. The model was implemented from scratch in C++, and developed within a Geographical Information Systems (GIS) environment that provides functions for the display and analysis of geographic informa tion. Finally, some spatial vaccination strategies were implemented which illustrates a useful purpose of this kind of simulation model, for instance for aiding in the definition of health policies. The documented simulations cover the sensitivity analysis of the parameters of the movement and infection model, and there are also simulations to reproduce a mumps outbreak that occurred in Portugal between 1993 and 1996. In this way, an application of the epidemic model was illustrated for a case study, that enabled an analysis and interpretation of the model efficiency.
| Type: | Thesis (Doctoral) |
|---|---|
| Title: | An agent-based approach to spatial epidemics through GIS |
| Identifier: | PQ ETD:593339 |
| Open access status: | An open access version is available from UCL Discovery |
| Language: | English |
| Additional information: | Thesis digitised by ProQuest. |
| URI: | https://discovery.ucl.ac.uk/id/eprint/1446013 |
Archive Staff Only
 |
View Item |
