This is a list of ideas that came out of a meeting between Nim and Ken on 21 July 2009.
- Seasonality. E.g. The flu season in the Northern Hemisphere is October to February. Also school and university terms and holidays could be modelled.
- Geography. Influences movement and contacts. Could use KML for Google Earth and other mapping services.
- Travel. Could model public transport, etc.
- Schools, work places, and households. Contacts within.
- Commuting to school and work.
- Age dependent differences in mixing and susceptibility.
- Mild or asymptomatic but still contagious infections. (Note that input data will typically miss these.)
- Could have each student model a community and then implement travel between them
- More interventions (e.g. school closing - when and for how long if at all)
- Replacement of one flu virus for another
- Risk of drug resistance depending upon how anti-viral drugs are administered
- Viruses becoming a new strain within a host
- Impacts on the economy (e.g. workers unable to work because infected or school closed or …)
- Modelling of sub-communities. E.g. At Oxford the colleges and departments
- Explicitly represent and visualise uncertainty (e.g. about the value of parameters)
Met again with Nim on 29 October 2009.
Talked about simpler version of his emergence model with 2 stages rather than 3 and something like a 1/4th probability for each change.
Could also model multiple virus infections and gene shuffling.
Rather than tag people as susceptible, infected, etc. they could have a possibly empty list of viruses that are infecting them. Viruses can be agents as well. Simplest to model exponential distribution for recovery — though there are ways to get a more realistic distribution.
In addition to building games where the player is a public health decision maker we talked about games from the viruses' point of view. I plan to attend this talk on this at the Royal Society: http://newsletters.royalsociety.org/c/1pVmRr4rp29N2FF
