Home | English | | | UGC | RGC
   
An International Hub for Academic Research
   
Analyses of Insurance Risk Models with Dividend Payments
   
Actuarial Science and Finance Model with Dependent Risks
   
Novel Molecular Mechanisms of innate host defense - Implications on periodontal health and disease
   
Avian flu nucleoprotein - from basic research to drug development
   
Effectiveness of Specialized Ventilation Systems in Hospitals of Hong Kong
   
Complex Network Models of Disease Propagation: Modelling, Predicting and Assessing the Transmission of SARS
   
RGC Collaborative Research fund - Layman Summaries of Projects Funded in 2008/2009 Exercise
   

       



In recent years several new infectious diseases have emerged: SARS, Avian Influenza and now Swine-Associated Influenza. In each case the challenge is how best to respond to these emerging threats in the face of incomplete information (concerning, for example, infectivity, transmission and relative risk to the community). Standard math-ematical models of disease propagation do a bad job of dealing with this lack of information as they focus on (unknown) properties of the disease. Nonetheless, as new diseases emerge we are faced with the challenge of how best to respond and to protect public health, while not over-reacting.

In this project we have applied new mathematical models from the field of complexity theory to the study of infectious diseases. These new methods focus on the properties of the community itself and incorporate whatever infor-mation concerning the infectious agent is available. As more information becomes available, the model can be improved. The basic tenet of complexity theory is that system of many simple, but interacting, elements can behave collec-tively in rather suprising and startling ways. To study these systems one must consider both the individual components and the ensemble behaviour of the entire system. In the application of these methods to infectious diseases we study how the fundamentally varied behaviour of the individuals within a community affects the spread of an infectious agent within that community. We apply both extensive computational modelling and techniques from statistical physics to describe the collective behaviour of the system.

In this completed project we applied these techniques to the spread of SARS in Hong Kong in 2003 and found several important results. In particular:

(i) the reported super-spreaders (single primary cases leading to a large number of secondary infections) can be explained by highly social (or highly connected) individuals – the disease itself may be no more virulent in these individuals than in the general community, and
 
(ii) the rate of spread of the disease in Hong Kong and the effect of the SARS outbreak can be largely attributed to spread within hospitals – proper infection control would have rendered the effect of the disease much less severe.

We have extended this work to the global spread of Avian influenza and found that the extremely dense connectivity between sites of infection means that the disease will naturally persist for a very long time. The only way to satisfactorily control the spread of Avian influenza in animals is with draconian culling measures (as implemented in Hong Kong in 1997), or by fundamentally altering the way in which live bird stocks interact (e.g. within distributed community-based wet markets).

We are now extending this work under a newly funded RGC competitive research grant titled "Complex network models of disease propagation: Heterogeneous analysis of propagation and control of SARS and Avian Influenza" and will focus on the combination of local and global community effects in the worldwide spread of disease. The primary focus of this ongoing work is Avian and now Swine-Associated Influenza, but the methods are applicable to any newly emerging contagion.


Dr. Michael Small
Department of Electronic and Information Engineering
The Hong Kong Polytechnic University
ensmall@polyu.edu.hk

Top