Transcript FetschPresentationQ4..

Smallpox Martyr Bio-terrorism
Modeling in Python
Computer Systems Lab 2010
Joe Fetsch
Purpose
What is the danger now?
Isn't smallpox gone forever?
Explanation of Purpose
Ok, so what are you doing to help?
Scenario
How are the terrorists going to attack?
Scenario, Cont'd
How will people react to that?
Why is that such a danger?
Scenario Cont'd
But that can't cause too many problems, right?
How could that possibly get worse?
Scenario Cont'd
But it's not over yet...
Scenario Cont'd
And it still gets worse.
Other Projects
Government simulations involving smallpox
exist, and government simulations involving
quarantines and vaccination exist, but not
both at the same time.
NetLogo Use

NetLogo was used to
develop a basic
understanding of the
disease modeling
system, but will not be
used to create the
smallpox model
NetLogo Virus Model
Smallpox

Child suffering from Smallpox
Much research was
done to fully
understand the
Variola virus in all
forms and its effects
on a population
Project Structure

Each dot represents
an Agent

The infection, after
Prodrome phase, will
then progress into a
more mature phase:

Ordinary

Modified

Malignant

Hemorrhaging

Confluent
Agent Movement

Agent Movement
Ignorance and
randomness
World Structure
•
Social Model
Effects on rate of
infection
Agents with few others
near them
Visual Representation




Green agents are
healthy
Yellow agents are in
early stage where not
contagious or visible
Orange agents are in
the prodromal phase,
exhibiting flu
symptoms
Red agents are
infected, contagious
Sugarscape-based model
Timeline


Research Smallpox
to understand
disease in order to
better implement in
program
Using NetLogo,
obtained a basic
understanding of
the model of
infection
Using Python,
created basic
model
Timeline
•
Developed a
model for
infection, hoping
to clarify my
values and prove
them accurate
with past data
Modeled fatality rates
Implemented
quarantine and
vaccination
possibilities
Testing

Man suffering from hemorrhagic
smallpox also known as black pox
– 100% fatal
Simulation has
begun, average
fatality rate in a city
around 35-40%
Still unknown:
•
Vaccination is unlikely
•
The chaos in the city would negate
military assistance for some time
•
Speculation
Simulations
•
Several different times of initiation for
vaccine and quarantine are used to
account for several different possible
scenarios.
Testing
•
Fatality rates are between 35 and 40%
•
All agents become infected within 6 or 7
months
•
90% of agents become infected within 4
and a half months
•
The graph on the next page shows the
results of many unhampered tests,
showing the population statuses over time
Quarantine
•
The quarantine
simulation shows a
world in which a
military quarantine has
isolated everyone from
each other.
Quarantine
•
The visual representation stops moving, yet
diseased agents continue to progress.
•
The line graph shows a red line at the time at which
the quarantine begins.
Description of the graph
In the graph above, the population of the city
has gone from 5000, the initial value, to 4052;
a fatality rate of 20%. However, the
population in this situation has been
quarantined after two months of the
simulation, while the rate of infection was still
increasing, which would lead to many more
cases of smallpox and many more fatalities.
Throughout the simulation, about half of the
agents became infected, which raises the
relative fatality rate to slightly less than 40%.
Quarantine Results
In the following slides, the results from calculating several
possible outcome times (30 days, 45 days, 60 days, and 75
days) ran 6 times to prevent outliers from significantly
affecting the results are shown.
The results before show the world immediately before the
quarantine, and the results after show the world 40 days
after, long enough to ensure that the remaining infected
agents will survive for the remainder of the simulation.
Therefore, the infected agents will be labeled immune.
Before the Quarantine
100%
90%
80%
% at each stage
70%
dead
immune
infected
prodrome
carriers
healthy
60%
50%
40%
30%
20%
10%
0%
30
45
60
Time (days) after attack
75
40 Days After Quarantine
Percent of population at each stage
100%
90%
80%
70%
60%
50%
dead
immune
healthy
40%
30%
20%
10%
0%
30
45
60
Beginning of Quarantine: Days after attack
75
Fatality/Infection Rates

The fatality rates for
quarantine simulations:

The infection rates for
the simulations:

30 days: 102 (2%)

30 days: 279 (6%)

45 days: 350 (7%)

45 days: 995 (20%)

60 days: 643 (13%)

60 days: 1766 (35%)

75 days: 992 (20%)

75 days: 2698 (54%)
Vaccine

The vaccine
simulation shows
a mass
distribution of
immunizations to
the smallpox
virus.
Vaccine
The visual representation continues moving, and
diseased agents continue to progress.
The line graph shows a green line at the time at
which the vaccine is distributed.
Vaccine Results
In the following slides, the results from calculating
several possible outcome times (60 days, 75 days, 90
days, and 105 days) ran 6 times to prevent outliers
from significantly affecting the results are shown.
The results before show the world immediately before
the vaccination, and the results after show the world
40 days after, long enough to ensure that the
remaining infected agents will survive for the
remainder of the simulation.
Therefore, the infected agents will be labeled immune.
Ring Vaccination
While ring vaccination has been successfully used
before to prevent the spread of smallpox, a
mass vaccination method is required because of
the nature of the first infection:
Because starting points for the infection are
1) spread out over a large area inside one city
and
2) at major transit locations (airports, etc)
The possibility to confine the infectious people
before they spread it further is very low.
Before Vaccination
100%
90%
80%
% at each stage
70%
60%
dead
immune
infected
prodrome
carriers
healthy
50%
40%
30%
20%
10%
0%
60
75
90
Time (days) after attack
105
After Vaccination
Percent of population at each stage
100%
90%
80%
70%
60%
50%
dead
immune
40%
30%
20%
10%
0%
60
75
90
Beginning of Vaccination: Days after attack
105
Fatality Rates
The fatality rates for the
vaccine simulations:
The infection rates for
the simulations:

60 days: 540 (11%)

60 days: 1747 (35%)

75 days: 991 (20%)

75 days: 2951 (59%)

90 days: 1247 (25%)

90 days: 3873 (77%)

105 days: 1396 (28%)

105 days: 3991 (80%)
Conclusions



Obviously, the earlier a vaccine is developed or
the earlier a quarantine is implemented, the
more lives can be saved.
However, this possibility for saving lives must
be weighed against the moral considerations of
confining people who may or may not want to
be confined.
We all know that there are going to be the
outliers who complain about being held against
their will... that's why I simulated the vaccine.
To Be Continued...
If further work was to be done, more data could be
gathered to create a more accurate and less
choppy graph of the expected results of the
experiment
Without taking real-life data (about five kinds of
illegal), very little information exists as to how
the disease actually spreads, and in order to
create a more specific scenario (school, etc),
much more understanding is required.