Outline of the course
Since its introduction Magnetic Resonance Imaging (MRI) offers the ability of image brain structure in-vivo. With the introduction in 1992 of functional MRI, the measure of brain structure has been extended to the in-vivo measure of brain function using the blood oxygen level dependent signal (BOLD). It is this combination (MRI/fMRI) that has become the preferred tool of neuroscientists to investigate non-invasively human brain structure and function. A part from applications to the understanding of basic sensory processing, cognitive functions they have been used in multiple clinical settings (e.g. Epilepsy, Schizophrenia) both for a better understanding of the biological basis of the disease and for the diagnosis (e.g. the detection of foci of epileptic discharges). This course provides basic introduction to the main anatomical (T1 weighted) and functional (T2* weighted; gradient-echo BOLD) contrast used in neuroscience. While describing basic acquisition concepts we will focus on post-processing strategies with particular interest on the statistical analysis of fMRI time series.
The course covers the following topics:
• Choosing the optimal anatomical sequences for use in cognitive neuroscience applications
• Prescribing functional data for optimal detection of targeted effects
• Design of an experiment in the fMRI environment
• Recognizing and preventing relevant artefacts that may prevent the fruitful analysis of imaging data
• Processing anatomical data with focus on inhomogeneity correction, spatial normalisation and segmentation (i.e. the detection of the white/grey matter boundary).
• Technologies to pre-process functional anatomical data with particular focus on slice-scan-time correction, motion correction, temporal and spatial filtering co-registration to anatomical data and spatial normalisation
• Analysis of the functional time series both at the single subject and group level in order to make inferences on the location activated by the chosen design.
Learning objectives
At the end of this course, students will have knowledge of:
• Basic concepts on anatomical and functional contrasts.
• Functional and anatomical sequences and their use.
• Typical artefacts in fMRI time series.
• Pre-processing strategies.
• Statistical analysis of fMRI time series.
• Functional localisation in volume and cortical surface space.
• Brain normalisation and analysis of group studies.
Content
fMRI experiments start with the setup of a stimulation protocol. The protocol has to be adapted to take into account the biophysics of the BOLD signal and constraints introduced by the fMRI environment. After the measurements two sets of data are generally available: 1. Anatomical images describing the structure (white-grey matter contrast) of the brain; 2. Functional time series representing the 4 dimensional representation of the brain activation in response to the stimulation protocol. Several strategies exist to analyse and combine these data in order to extract information on the experimental design. Pre-processing strategies have been developed to remove effects of non interest and increment statistical power. The end result of a typical analysis pipeline is the superimposition of functional activation maps (i.e. statistical maps describing regions activated by the experimental manipulation) on anatomical data. The latter can be represented in volumetric fashion (transversal, sagittal and coronal view) or can be used to extract surfaces representing the white-grey matter border.
This course will treat the basic analysis pipeline, starting with stimulus design, and experimental setup and ending with pre-processing and statistical analysis.
Overview of tasks and lectures
There will be 10 lectures of 2 hours distributed over 5 days.
• Introduction to MRI and the basic anatomical contrasts.
• Pre-processing of anatomical data.
• The fMRI signal basic biophysics concepts.
• fMRI imaging sequences: setup and applications.
• Experimental designs: block, event related and fast even related.
• Pre-processing of functional data.
• Statistical analysis of imaging data I: single subjects.
• From single subjects to group analysis: spatial normalization and inter subject alignment.
• Statistical analysis of imaging data II: group analysis.
• Summary
Position within the programme
This course gives a basic overview of functional and anatomical MRI data analysis. A standard pipeline of data processing will be detailed and will serve as introduction for more advanced data analysis courses. Knowledge of basic signal processing concepts (e.g. filtering) is sufficient to attend the course.
Teaching format
Structure
The course is a one week-long residential module consisting of 10 meetings of 2 hours. In addition, each day the students will perform hands-on exercises in data acquisition and analysis of anatomical and functional data guided by tutors.
Grading
Passing the course requires an 85% attendance to the lectures and practical sessions. Students will report on the post-course assignments in written form which will be evaluated by the course coordinator(s).