Title: The%20Basics%20of%20fMRI
1The Basics of fMRI
- Overview
- Physiological origins (BOLD)
- Physical origins (MRI)
- Physical measurements of BOLD
- Paradigm (task) design
- Analysis of task-activated fMRI
- Resting fMRI
2Physiological origins
stimulus
Change in oxydeoxy ratio
Neurophysiological coupling neuronal activity ?
blood oxygenation
Logothetis et al. (2001) Nature 412 150 LFP
local field potentials MUA multi unit
activity SDF spike density function
3Physiological origins
- Local hemodynamic changes
- ROY, C. W., SHERINGTON, C. S.
- On the regulation of the blood-supply of the
brain. - J. Physiol. (Lond.) 11 85-108, 1890.
- Increase in local blood flow (50)
- Increase in local blood volume
- Small increase in oxygen consumption (15)
- Increased flow means reduced O2 extraction
hence oversupply of blood - Haemodynamic response function (HRF)
4Physical origins
Structural MRI
Proton (1H) has a magnetic moment which can be
non-toxically manipulated with transitory
magnetic fields (B1) and RF energy to produce
images of its local environment.
5Physical origins
- Amplitude of signal determined by
- Proton density more protons, more signal
- T1 (spin-lattice relaxation) Stimulated by local
magnetic field fluctuations due to magnetic
properties of other molecules. - T2 Enhanced dephasing due to inhomogeneities in
local magnetic field (inc. B0) susceptibility. - T2 (spin-spin relaxation) residual dephasing
during dephase-rephase period - Contrast agents changing local susceptibility
6Physical origins
- Definitions
- Time to Repetition (TR) The TR is the time
between consecutive sequence initiations - Time to Echo (TE) TE determines the sensitivity
to T2, which varies for different tissues - Acquisition Time (TA) The time between
acquisitions - (TR gt TE often TRTA)
7Detecting BOLD
Deoxyhaemoglobin is paramagnetic, positive
susceptibility
- Changes local susceptibility and therefore T2
- MRI Sequence requirements
- T2 sensitive
- fast
- whole brain
- gt Echo planar imaging (EPI)
Thulborn et al., 1982
8MRI Safety
- Powerful magnetic field
- Extreme forces during rapidly changing gradients
- RF energy deposition
- Confined environment / restraints
- Need for controlled access and screening
9Paradigm Design
Block paradigm
Event Related
Event Related (compressed)
10Paradigm Design
- Magnitude of the BOLD effect can be modulated
experimentally - Task difficulty (set size, encode-retrieve
delay, stimulus frequency, stimulus load,
distracters) - Learning (decrease in activation)
- Accommodation (decrease in activation)
- Other stimulus (pharmaceutical, mood induction,
age) - somatosensory internal standard, on-line
behavioural data - Baseline tasks must be appropriate for cognitive
subtraction
11Data Analysis
- EPI MRI data volumes are continuously acquired
whilst the subject performs some cognitive task
(paradigm). - Following pre-processing, analysis proceeds as
- The within-group activation engendered by the
paradigm on average - The between-group difference in activation or the
within-group correlation of activation with some
other variable
12Data Analysis summary
- Pre-process data from each individual to correct
subject motion - Estimate response at each voxel (General linear
model) - Map subjects into same anatomic space
- Statistically infer activation for each group
- Statistically infer difference between groups
13Data Analysis of individuals
Response estimation (GLM)
Difference between male/female Europeans
Difference between European/Japanese males
Subject movement
Spatial normalisation
14Real time fMRI
Block paradigm Two conditions (motor,
spatial) Two distinct regions Infer responses
Monti et al, 2010
15Data Analysis of groups
Differences extend outside activated network
Within-group H0 activation is uncorrelated to
stimulus
Between-group H0 zero mean difference between
groups (whole brain)
16Resting state fMRI
Doing nothing
Seed-based correlations
Deactivations regions with greater activations
in task-absent conditions
17Resting state fMRI
18End