Probabilistic Graphical Models

Code:

OPT24

Acronym:

MGP14

Keywords
Classification	Keyword
OFICIAL	Medicine

Instance: 2017/2018 - 1S

Active?	Yes
Responsible unit:	Department of Community Medicine, Information and Health Decision Sciences
Course/CS Responsible:	Integrated Master in Medicine

Cycles of Study/Courses

Acronym	No. of Students	Study Plan	Curricular Years	Credits UCN	Credits ECTS	Contact hours	Total Time
MIMED	17	Mestrado Integrado em Medicina- Plano oficial 2013 (Reforma Curricular)	2	-	1,5	14	41
			3
			4

Teaching language

Suitable for English-speaking students

Objectives

This unit aims to empower the students with necessary knowledge and skills to use modern methods of probabilistic reasoning for biomedical problems, more specifically regarding theory and practice of Bayesian networks for interdependencies exploration and clinical decision support.

Learning outcomes and competences

• Identify different types of uncertainty inherent in clinical practice.


• Explore the main objectives of Bayesian inference for clinical decision support.


• Describe the theory of Bayesian networks, its objectives and the main features.


• Observe clinical applications of Bayesian networks in different areas and outcomes.

Working method

Presencial

Program

1. Introduction:

1.1 Motivation and examples;

1.2 Probability and medical applciations;

1.3 Probabilistic graphical models;

1.4 Introduction to Bayesian networks.

1. Bayesian networks:

2.1 Semantics and factorization;

2.2 Probabilistic influence flow;

2.3 Conditional independence andnaive Bayes;

2.4 Causal independence.

1. Bayesian networks in clinical research and decision support

3.1 Diagnosis

3.2 Prognosis

3.3 Etiology and risk

1. Building Bayesian networks from data

4.1 Machine learning

4.2 Bayesian network parameter estimation

4.3 Bayesian network structure learning

4.4 Learning and inferring from incomplete data.

Mandatory literature

Darwiche, A.; Modeling and Reasoning with Bayesian Networks, Cambridge University Press, 2009
Cowell, R.G., Dawid, P., Lauritzen, S.L., Spiegelhalter, D.J.; Probabilistic Networks and Expert Systems, Springer Verlag, 1999
ucas, P. J. F., van der Gaag, L. C., & Abu-Hanna, A.; Bayesian networks in biomedicine and health-care , Artificial intelligence in medicine, 2004
Darwiche, A.; Bayesian networks, Communications of the ACM, 2010 (53(12), 80–90)
Sierra, B., & Larranaga, P.; Medical Bayes Networks In International Symposium on Medical Data Analysis, Springer Verlag, 2000 (pp. 4–14)

Teaching methods and learning activities

Theoretical lectures and practical lessons, with topic discussion, individual and group exercises, and hands-on training with proper software. Evaluation will be based on assignments and final exam.

Software

R
SamIam

keywords

Physical sciences > Mathematics > Statistics
Physical sciences > Mathematics > Probability theory
Physical sciences > Mathematics > Statistics > Medical statistics

Evaluation Type

Distributed evaluation with final exam

Assessment Components

Designation	Weight (%)
Exame	40,00
Trabalho escrito	60,00
Total:	100,00

Calculation formula of final grade

Evaluation will be based on (60%) assignments and final exam (40%).