Pandemic Modeling
and Ethics of Care
Carol Cuesta-Lazaro
![](https://s3.amazonaws.com/media-p.slid.es/uploads/993552/images/9859021/pasted-from-clipboard.png)
IAIFI Fellow - Institute for Artificial Intelligence and Fundamental Interactions (MIT)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/993552/images/9201524/pasted-from-clipboard.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/993552/images/9201514/pasted-from-clipboard.png)
Medical Imaging
Agent Based simulations
Observed
Simulated
Cosmology
Astrophysics
Simulations
HPC
Scientific question
Statistics ML
![](https://s3.amazonaws.com/media-p.slid.es/uploads/993552/images/6926154/pasted-from-clipboard.png)
The JUNE epidemiological model
![](https://s3.amazonaws.com/media-p.slid.es/uploads/900381/images/7383584/pasted-from-clipboard.png)
June Dalziel Almeida
![](https://images.idgesg.net/images/article/2018/06/digital-twins_woman-in-profile_ai_mirror_duplicate_duo_pair-100760562-large.jpg?auto=webp&quality=85,70)
A virtual society
![](https://d1e00ek4ebabms.cloudfront.net/production/55506c34-3a3d-456e-8ea8-cf71370708b4.jpg)
Viral transmission
How does an epidemic propagate through a population?
Why does it affect some sectors of the population more than others?
How do we best stop it?
The importance of complex models
![](https://s3.amazonaws.com/media-p.slid.es/uploads/900381/images/8955645/pasted-from-clipboard.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/993552/images/9685082/pasted-from-clipboard.png)
- Abstraction (and avoiding biases!) can sometimes be difficult, due to caring for your the subjects of your research
- Trading speed for correctness when timeliness matters
- Communicating effectively is as important as doing the right work
What I learned from JUNE that I couldn't have learned from Astrophysics
![](https://s3.amazonaws.com/media-p.slid.es/uploads/993552/images/10349417/Screenshot_2023-03-29_at_1.07.40_PM.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/993552/images/10349416/Screenshot_2023-03-29_at_1.11.40_PM.png)
Limited understanding
1. How the virus behaves
2. How effective are different preventive measures
3. What is the current state?
The challenges of complex detailed models
![](https://s3.amazonaws.com/media-p.slid.es/uploads/900381/images/9466637/pasted-from-clipboard.png)
Find the model under which the data is most likely
![](https://s3.amazonaws.com/media-p.slid.es/uploads/993552/images/9685112/pasted-from-clipboard.png)
There isn't one single model that explains reality
Uncertainty
Goal: Provide a tool for understanding the impact of operational interventions during the COVID-19 crisis in constrained and difficult operational contexts (e.g. Cox’s Bazar) based on data, alerts and evidence to aid in decision making
![](https://s3.amazonaws.com/media-p.slid.es/uploads/993552/images/9685090/pasted-from-clipboard.png)
Adapting JUNE to refugee settlements
Cox's Bazar
-
Largest settlement in the world
-
In some areas, the settlement is denser than New York City (44,000 people estimated to be living per square kilometer)
-
High risk of COVID transmission
![](https://s3.amazonaws.com/media-p.slid.es/uploads/993552/images/9668661/pasted-from-clipboard.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/993552/images/9668689/pasted-from-clipboard.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/993552/images/9668690/pasted-from-clipboard.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/993552/images/9669037/pasted-from-clipboard.png)
Scenario modelling
Challenges
Building a digital twin, data?
biases?
Monitoring: Infections, hosptilizations and deaths?
DATA
Access to health care?
Model assumptions and cultural differences
Need for more creative policies
POPULATION DIFFERENCES
deck
By carol cuesta
deck
- 120