|
| Re: Malaria [message #529755 is a reply to message #529691] |
Tue, 27 April 2010 06:43   |
 Arik Kershenbaum
Messages: 13
Registered: January 2010
Location: Israel |
Junior Member |
|
|
Much as I'd like to start doing zoonosis work in STEM, I'm not sure that malaria is the disease to start with. Since it's so widely endemic, changes and variations in incidence are more to do with varying prevalence and differences in mosquito ecology, rather than flow of infection. We wouldn't be leveraging the ability of STEM to model movement of infected individuals. WNV is possibly a more promising first step.
Having said that, we certainly could use STEM (even if not to its full capability) to model the effect of anti-malarial drugs. I've never looked at the innoculator code; how comprehensive is the support for medical intervention? Can we model a fixed rate of innoculation (or oral medication) in a particular region for example?
Another interesting direction we could take the example of malaria, is to look at the effect of movement of infected individuals (e.g. urbanisation), assuming that vector ecology is locally constant. I think we would need two things for this. (1) The ability to maintain a fixed incidence/prevalence in certain administrative regions, representing the endemic rural areas, and (2) variable transmission in different administrative regions, representing differences in vector ecology in rural and urban areas. That way, we could see under what conditions an influx from the villages to the cities of carriers of the parasite would lead to endemic establishment of the disease in urban areas.
I suspect that neither of these two features is supported; am I correct?
|
|
|
|
| Re: Malaria [message #530175 is a reply to message #529755] |
Wed, 28 April 2010 18:12 |
Stefan Edlund
Messages: 127
Registered: July 2009
Location: IBM |
Senior Member |
|
|
To try and answer some of Arik's questions:
I've never looked at the innoculator code; how comprehensive is the support for medical intervention? Can we model a fixed rate of innoculation (or oral medication) in a particular region for example?
The code you are referring to is the Trigger code in STEM. It's the ability to say after a condition is true, modify some aspect of the simulation. Right now, it's not as advanced as we would like it to be. A group in India used STEM to invoke an inoculator after a certain time had progressed in the simulation, and that's about all we can do right now. The inoculation only happens once, so for instance at day 20, x % of the population is vaccinated. It would probably be more useful to use a daily rate instead.
Another interesting direction we could take the example of malaria, is to look at the effect of movement of infected individuals (e.g. urbanisation), assuming that vector ecology is locally constant. I think we would need two things for this. (1) The ability to maintain a fixed incidence/prevalence in certain administrative regions, representing the endemic rural areas, and (2) variable transmission in different administrative regions, representing differences in vector ecology in rural and urban areas. That way, we could see under what conditions an influx from the villages to the cities of carriers of the parasite would lead to endemic establishment of the disease in urban areas.
I suspect that neither of these two features is supported; am I correct?
This is a good suggestion. I think we can handle movement of people between rural and urban areas simply by allowing the fraction between the two vary between simulations and exploring the effect (like we do for other parameters today). On the other hand, if you want to model movement taking place within a single simulation, we need to improve the migration code in STEM. We do have the concept of migration edges, but they are not really being used currently.
|
|
|
|
|
|
|
|
| Re: Malaria [message #552962 is a reply to message #529691] |
Sun, 15 August 2010 08:10 |
steph 
Messages: 7
Registered: August 2010 |
Junior Member |
|
|
Philippines.
This is a college SP and using STEM is a proposal. I'm still not sure if it will be approved of.
It doesn't have to be grand. There aren't very many vector host model simulations available. Modeling would be the focus of the work. Its interesting because most simulations involve human contact. This is something different.
by the way, when you say creating a model, do you mean
creating one as a 'user of STEM' or as a 'programmer of STEM'?
[Updated on: Mon, 16 August 2010 05:46] Report message to a moderator |
|
|
| Re: Malaria [message #553016 is a reply to message #552869] |
Mon, 16 August 2010 06:09 |
steph
Messages: 7
Registered: August 2010 |
Junior Member |
|
|
I don't know if I can be of any help, after all Ive only just seen STEM this Friday. I haven't explored it yet. Also, I have a class on Wed.
This is very interesting, though I admit that my work doesn't have to be Malaria, I'm just interested in Vector Host Model and ways in representing the disease transfer.
Also, I'm not sure if ill be allowed to do it(the work) in STEM. But like what Mr. Kershenbaum said, modeling of the disease transfer can be done in a multitude of ways, In fact, Networking(or something like it) has been used before but that is with same species transfer (usually STD's).
I just think that essentially, it has to be done with two populations and finding a way to connect them(like connecting layers in GIS).
|
|
|
|
| Re: Malaria [message #563459 is a reply to message #529691] |
Tue, 27 April 2010 06:43 |
Arik Kershenbaum
Messages: 13
Registered: January 2010
Location: Israel |
Junior Member |
|
|
Much as I'd like to start doing zoonosis work in STEM, I'm not sure that malaria is the disease to start with. Since it's so widely endemic, changes and variations in incidence are more to do with varying prevalence and differences in mosquito ecology, rather than flow of infection. We wouldn't be leveraging the ability of STEM to model movement of infected individuals. WNV is possibly a more promising first step.
Having said that, we certainly could use STEM (even if not to its full capability) to model the effect of anti-malarial drugs. I've never looked at the innoculator code; how comprehensive is the support for medical intervention? Can we model a fixed rate of innoculation (or oral medication) in a particular region for example?
Another interesting direction we could take the example of malaria, is to look at the effect of movement of infected individuals (e.g. urbanisation), assuming that vector ecology is locally constant. I think we would need two things for this. (1) The ability to maintain a fixed incidence/prevalence in certain administrative regions, representing the endemic rural areas, and (2) variable transmission in different administrative regions, representing differences in vector ecology in rural and urban areas. That way, we could see under what conditions an influx from the villages to the cities of carriers of the parasite would lead to endemic establishment of the disease in urban areas.
I suspect that neither of these two features is supported; am I correct?
--
==============
Arik Kershenbaum
Prof Blaustein's laboratory
Department of Evolutionary and Environmental Biology
University of Haifa
http://research.haifa.ac.il/~leon/html/Arik_Page.htm
|
|
|
|
| Re: Malaria [message #563552 is a reply to message #563459] |
Wed, 28 April 2010 18:12 |
Stefan Edlund
Messages: 127
Registered: July 2009
Location: IBM |
Senior Member |
|
|
To try and answer some of Arik's questions:
I've never looked at the innoculator code; how comprehensive is the support for medical intervention? Can we model a fixed rate of innoculation (or oral medication) in a particular region for example?
The code you are referring to is the Trigger code in STEM. It's the ability to say after a condition is true, modify some aspect of the simulation. Right now, it's not as advanced as we would like it to be. A group in India used STEM to invoke an inoculator after a certain time had progressed in the simulation, and that's about all we can do right now. The inoculation only happens once, so for instance at day 20, x % of the population is vaccinated. It would probably be more useful to use a daily rate instead.
Another interesting direction we could take the example of malaria, is to look at the effect of movement of infected individuals (e.g. urbanisation), assuming that vector ecology is locally constant. I think we would need two things for this. (1) The ability to maintain a fixed incidence/prevalence in certain administrative regions, representing the endemic rural areas, and (2) variable transmission in different administrative regions, representing differences in vector ecology in rural and urban areas. That way, we could see under what conditions an influx from the villages to the cities of carriers of the parasite would lead to endemic establishment of the disease in urban areas.
I suspect that neither of these two features is supported; am I correct?
This is a good suggestion. I think we can handle movement of people between rural and urban areas simply by allowing the fraction between the two vary between simulations and exploring the effect (like we do for other parameters today). On the other hand, if you want to model movement taking place within a single simulation, we need to improve the migration code in STEM. We do have the concept of migration edges, but they are not really being used currently.
|
|
|
|
|
|
|
|
|
| Re: Malaria [message #564696 is a reply to message #564639] |
Mon, 16 August 2010 06:09 |
steph
Messages: 7
Registered: August 2010 |
Junior Member |
|
|
I don't know if I can be of any help, after all Ive only just seen STEM this Friday. I haven't explored it yet. Also, I have a class on Wed.
This is very interesting, though I admit that my work doesn't have to be Malaria, I'm just interested in Vector Host Model and ways in representing the disease transfer.
Also, I'm not sure if ill be allowed to do it(the work) in STEM. But like what Mr. Kershenbaum said, modeling of the disease transfer can be done in a multitude of ways, In fact, Networking(or something like it) has been used before but that is with same species transfer (usually STD's).
I just think that essentially, it has to be done with two populations and finding a way to connect them(like connecting layers in GIS).
|
|
|
|
Powered by
FUDForum. Page generated in 0.05082 seconds
] 
Search
Help
Register
Login
Home
