Course ObjectiveThe course treats the theory of parameter estimation problems in
general, but the theory is illustrated extensively by examples from
medical and biological sciences and brain imaging (fMRI and MEG/EEG) in
particular. Linear and non-linear regression analysis is treated, as
well as confidence intervals and significance testing. The goal of the
course is to provide insight into the theory of parameter estimation and
to develop a critical attitude towards its application and
interpretation in order to avoid inconsistent and improper use of the
Course ContentLinear-non linear parameter models, basic matrix-vector algebra,maximum
likelihood principle, correlated-uncorrelated noise, OLS, GLS, data
outliers, nuisance parameters, linear (time invariant) filters,
projection filters t-test, F-test, confidence intervals, multiple
testing, fMRI data model, missing data, MEG/EEG source localisation.
These topics are treated in the form of a series of lectures alternated
Extra topics: L1 en L2 norms.
Teaching MethodsLecture and optional (MatLab) exercises.
Method of AssessmentWritten exam plus bonus point for critical review of scientific paper.
Entry RequirementsIn order to participate in this course, students are required to have
successfully completed Image Processing for MNS (X_422612) and Medical
Imaging for MNS (XM_0063).
LiteratureA syllabus and slides will be provided by the lecturer.
Recommended background knowledgeBasic Matlab experience is recommended to enable completion of optional
|Language of Tuition||English|
|Faculty||Faculty of Science|
|Course Coordinator||dr. I.H.M. van Stokkum|
|Examiner||dr. J.C. de Munck|
dr. J.C. de Munck
dr. I.H.M. van Stokkum
You need to register for this course yourself
Last-minute registration is available for this course.
|Teaching Methods||Lecture, Practical|
This course is also available as: