In this three-day short course, we will develop agent-based models for simulating a variety of biomedical problems. We will be using the NetLogo software package, which is in extensive use in agent-based modeling of a very wide variety of domains, not solely biomedical. NetLogo supports an intuitive graphical interface that allows the user to change parameters that have been specified by a developer, as well as a programming language (based on Logo, and early entry into the game programming market). Although NetLogo will be used for purposes of creating the models, we will also introduce the theory of agent-based and complex systems as well as a process for modeling agent-based models that has a seamless development pipeline from initial specification of a problem domain through the implementation of a model.

Dal 08/07/2019 al 10/07/2019
09:30 – 17:00

Day 1: General Introduction to agent-based modeling and NetLogo
Morning Course overview and introduction
Agent-based modeling (Ch. 1)
The NetLogo platform (Ch 2.)
Example: AIDS
Afternoon Creating an ABM with NetLogo
The ODD modeling framework (Ch. 3)
The NetLogo programming language (Ch. 4)
Example: AIDS model, found in the NetLogo Models Library, Sample Models/Social Science folder.
Exercise: Implementing a simple model: Food availability
Reading: Chiaccio F, et al.: Agent-based modeling of the immune system: NetLogo, a
promising framework. BioMed Research International, Article ID 907171, 2014.

Day 2: Contagion and infectious disease modeling
Morning Model design concepts I
Emergence (Ch. 7-8)
Observation (Ch. 9)
Sensing (Ch. 10)
Adaptive behavior (Ch. 11)
Afternoon Model design concepts II
Prediction (Ch. 12)
Interaction (Ch. 13)
Scheduling (Ch. 14)
Stochasticity (Ch. 15)
Collectives (Ch. 16)
Example: http://modelingcommons.org/browse/one_model/4332#model_tabs_browse_procedures
Exercise: Simulating an influenza outbreak using a common contagion model
Reading: Cooley P, et al.: The role of subway travel in an influenza epidemic: A New York City simulation. Journal of Urban Health: Bulletin of the New York Academy of Medicine 88(5):982-994, 2011.

Day 3: Built-environment interactions
Morning Pattern-oriented matching
Patterns for model structure (Ch. 17-18)
Theory development (Ch. 19)
Parameterization and calibration (Ch. 20)
Afternoon Model analysis and course wrap-up
Analyzing and understanding agent-based models (Ch. 21-22)
Sensitivity, uncertainty and robustness (Ch. 23)
Incorporating geography in an ABM
What next? (Ch. 24)
Example: Pollution model, found in the NetLogo Models Library, Sample Models/Urban Suite folder.
Exercise: Environmental modeling with NetLogo
Reading: Yang Y and Diez-Roux AV: Using an agent-based model to simulate children’s active travel to school. International Journal of Behavioral Nutrition and Physical Activity 10(67): 1-8, 2013.

Credits are available for students of the PhD program in HEALTH TECHNOLOGIES, BIOENGINEERING AND BIOINFORMATICS

Room D8, D Floor; Room C1, C Floor;
Dept. for Electrical, computer and biomedical engineering.
Via Ferrata, 5 – Pavia