Pattern detection in biomedical data
1
2017-2018
03019579
Biomedical Engineering
Portuguese
English
Face-to-face
SEMESTRIAL
6.0
Elective
3rd Cycle Studies
Recommended Prerequisites
Basic programming skills
Teaching Methods
Lecturing, discussing and resolving practical problems.
Learning Outcomes
The course intends to apply Data Science principles to the analysis of biomedical data. The expected learning outcomes are:
* Adopt the best principles of reproducible research, both for analysing data and communicating results
* Understand the difference between raw data and data suitable for statistical analysis and gain the competences to be able to extract the latter from the former
* To analyze data using appropriate statistical procedures
* Acquire the skills to propose and assess models for medical data
* To use suitable software tools
* To judiciously evaluate the results of studies published in the literature
Work Placement(s)
NoSyllabus
* Getting and cleaning biomedical or clinical data
* Introduction to reproducible research
* Exploratory data analysis: normalization of quantitative variable, dichotomisation of qualitative variables. Detecting and handling outliers. Univariate and multivariate graphical analysis; techniques to display the information within variables, cluster analysis and data reduction
* Statistical prediction, classification and regression methods
* Supervised learning statistical methods. Building ensembles of classifiers.
Head Lecturer(s)
Francisco José Santiago Fernandes Amado Caramelo
Assessment Methods
Assessment
Project: 100.0%
Bibliography
Análise Estatística, com utilização do SPSS; João Maroco, Edições Silabo;
Fundamentals of Biostatistics, Bernard Rosner, Thomson Brooks/Cole, 2006
Bioestatística, Epidemiologia e Investigação, A. Gouveia de Oliveira, Lidel
Métodos Quantitativos em Medicina, Massad, Menezes, Silveira & Ortega ed. Manole, 2004
Pattern Classification; Richard Duda, Peter Hart, David Stork; John Wiley & Sons, Inc
An Introduction of Support Vector Machines; Nello Christianini, John Shawe-Taylor; Cambridge University Press