Information diffusion, network overlap and the modelling of epidemics
Lead Research Organisation:
University of Sussex
Department Name: Sch of Mathematical & Physical Sciences
Abstract
Many mathematical models of infectious disease spread assume a 'passive' population where individuals' probability of becoming infected and their contact pattern is not altered in any way by the presence of the disease. However, in a realistic scenario, due to the diffusion of the information triggered by the disease, individuals within a population react and can lower their probability of becoming infected. Therefore, disease transmission and the diffusion of information are interlinked and need to be considered simultaneously. In this project, we propose to develop and to analyse a class of novel mathematical and simulation models that allow us to capture the concurrent spread of the disease and information that is generated by the presence of the disease.The project will start by using a simple model formulation in terms of differential equations (i.e. pair-approximation models) to understand and uncover key interactions between the parallel spread of the disease and information. Building on this, we will generalize results for networks of arbitrary size and consider the case when disease and information spread through networks of contacts that are completely overlapping.Using a continuous Markov chain formulation, we will derive analytical results on simple model networks and determine conditions needed to reduce or stop disease transmission. By using network overlap or similarity measures from graph theory, we will adapt and develop custom made overlap measures that are meaningful when disease spread and the diffusion of information are concurrently considered.
Organisations
People |
ORCID iD |
| Istvan Kiss (Principal Investigator) |
Publications
Hartley C
(2014)
Identification of Criticality in Neuronal Avalanches: II. A Theoretical and Empirical Investigation of the Driven Case.
in Journal of mathematical neuroscience
Hatzopoulos V
(2011)
Multiple sources and routes of information transmission: Implications for epidemic dynamics.
in Mathematical biosciences
Kiss I
(2012)
Modelling approaches for simple dynamic networks and applications to disease transmission models
in Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences
Kiss IZ
(2010)
The impact of information transmission on epidemic outbreaks.
in Mathematical biosciences
Simon P
(2012)
From exact stochastic to mean-field ODE models: a new approach to prove convergence results
in IMA Journal of Applied Mathematics
Simon PL
(2011)
Exact epidemic models on graphs using graph-automorphism driven lumping.
in Journal of mathematical biology
Taylor M
(2012)
From Markovian to pairwise epidemic models and the performance of moment closure approximations.
in Journal of mathematical biology
| Description | Prevention and/or control of many infectious disease outbreaks rely on the transmission of information via either simple social interactions between individuals or targeted campaigns at population level. Hence, individuals are not 'passive' players, and when faced with an outbreak they will try to avoid becoming infected and/or passing infection on. However, taking such measures relies on the fast and efficient transmission of information about the disease as well as the assimilation and interpretation of it by single or groups of individuals. It is in general very unlikely that routes of disease transmission overlap with routes of disease transmission and thus it is important to understand how the overlap or lack of it between the two networks of contacts impacts on the outcome of the epidemic. More importantly information can be transmitted via different routes (e.g. individual-to-individual, large campaigns, media etc) and the effectiveness of various routes or combinations of these are not at all easy to establish. Using mathematical models, in this project we shed some light on how the two processes interact and interpreted our findings in terms of the design of effective disease prevention campaigns and epidemic control measures. In this project with tackled this problem by developing two mechanistic mathematical models that account for the concurrent spread of a disease and the information generated by its presence. A) First, we developed a compartmental model where individuals are classified both with respect to disease (i.e. susceptible or infected) and information (i.e. having or not having the information). The model accounted for both information generation and transmission and has allowed us to show that fast dissemination of information can lead to disease eradication. Where eradication is not possible, the transmission of information is still beneficial as it leads to lower prevalence of infection in the population. B) Second, we extended the previous simple model to pair-wise and simulation models that are much better suited to capturing the interaction of the individuals at the local level and to explore the effect of overlap between disease and information transmitting contacts. We found that: a. Contact based transmission of information is the most efficient as it creates multiple secondary sources of information and outcompetes processes such as the global transmission of information. b. The effect of neighborhood overlap is tightly linked to the primary generation of information (e.g. information about the disease arises directly from infected individual or from those already in treatment) and can result in the creation of different correlation structures in responsiveness and also between responsiveness and infection. c. The mixing patterns in information can significantly alter the basic reproductive number (R0) of the disease. This is an important result since it opens up the possibility to tune or optimize responsiveness in order to limit the potential of new outbreaks. |
| Exploitation Route | This a relatively new research field and recent results, including my own results, were incorporated in a recent book "Incorporating human behaviour in epidemic dynamics: a modelling perspective"; in Modelling the interplay between human behaviour and the spread of infectious diseases (2013) New York: Springer. ISBN 9781461454731, 9781461454748. |
| Sectors | Education,Healthcare |
| Description | The main impact thus far was mainly academic or influencing further research. Most notably the results from my research are now part of a specialist book in the newly emerging research area of coupling human behaviour with epidemic models - see "Incorporating human behaviour in epidemic dynamics: a modelling perspective" in Modelling the interplay between human behaviour and the spread of infectious diseases (2013) New York: Springer. ISBN 9781461454731, 9781461454748. The other impact is via a JAVA software for the visualisation of networks and epidemic spread which is currently used to develop outreach talks/lessons to be delivered to local schools. The lesson plans are ready to be tested by delivering them to first year undergraduate mathematics students, see the Software and Technical Products entry. |
| First Year Of Impact | 2014 |
| Sector | Education |
| Impact Types | Societal |
| Description | IMA Small Grant Scheme |
| Amount | £425 (GBP) |
| Funding ID | SGS01/13 |
| Organisation | National institute for Polar Research |
| Sector | Academic/University |
| Country | Japan |
| Start | 06/2013 |
| End | 07/2013 |
| Title | Network and epidemic visualisation software |
| Description | We developed a custom-made, user-friendly simulation package that allows the real-time visualisation of disease transmission on static and dynamic networks. This hands-on simulation tool has a wealth of features that allows users to select different network architectures, to simulate the spread of diseases with different characteristics, and to test various control scenarios. This tool can be used during outreach activities (e.g. master classes, science fairs/festivals and other extra-curricular activities) to actively engage audiences and to demonstrate the usefulness of mathematics and mathematical modelling. Furthermore, this tool can be used in research to gain intuitive understanding of more complex processes. The software was developed using Java. |
| Type Of Technology | Software |
| Year Produced | 2011 |
| Open Source License? | Yes |
| Impact | Currently, 2014/2015, the Department of Mathematics at Sussex is in the process of developing two outreach lessons where this software is actively used. The lesson plans are in the testing phase. |
| URL | https://github.com/mbatchkarov/Plovdiv |
| Description | Invited Research Talk |
| Form Of Engagement Activity | A talk or presentation |
| Part Of Official Scheme? | No |
| Geographic Reach | National |
| Primary Audience | Participants in your research and patient groups |
| Results and Impact | Departmental research seminar at University of Sterling. Mainly academic/research. |
| Year(s) Of Engagement Activity | 2010 |
| Description | Multiple sources and routes of information transmission: Implications for epidemic dynamics (Invited talk), International Workshop on Amorphous Computing and Complex Biological Networks, University of Sheffield, 17-20 August 2010. |
| Form Of Engagement Activity | A formal working group, expert panel or dialogue |
| Part Of Official Scheme? | No |
| Geographic Reach | International |
| Primary Audience | Participants in your research and patient groups |
| Results and Impact | Discussed challenges, new research directions/approaches/methods. Further model development and working towards a book on the topic of coupling behavioural response with epidemic models. |
| Year(s) Of Engagement Activity | 2010 |
| Description | Multiple sources and routes of information transmission: Implications for epidemic dynamics (Invited talk), Mini-symposia on Information, human behaviour and disease), 8th European Conference on Theoretical and Mathematical Biology, Krakow, June 2011 |
| Form Of Engagement Activity | A talk or presentation |
| Part Of Official Scheme? | No |
| Geographic Reach | International |
| Primary Audience | Participants in your research and patient groups |
| Results and Impact | Presenting to other experts in the field. Being invited to contribute to a specialist book on the topic of the Interplay Between Human Behaviour and the Spread of Infectious Diseases. |
| Year(s) Of Engagement Activity | 2011 |
| Description | Special Session on Evolution equations and mathematical biology, 8th AIMS Conference on Dynamical Systems, Differential Equations and Applications, Dresden, 25-28 May 2010. |
| Form Of Engagement Activity | A talk or presentation |
| Part Of Official Scheme? | No |
| Geographic Reach | International |
| Primary Audience | Participants in your research and patient groups |
| Results and Impact | I was invited to participate in a dedicated conference session. Mainly academic, sharing ideas/methods/open problems etc. |
| Year(s) Of Engagement Activity | 2010 |