Introduction to fMRI Analysis

1
Introduction to fMRI Analysis
Kennedy Center Training Series April 18,
2007 Baxter P. Rogers
2
Basics of fMRI
3
MRI vs. fMRI
Functional MRI (fMRI) studies brain function.
MRI studies brain anatomy.
4
MRI vs. fMRI
MRI
fMRI
high resolution (1 mm)
low resolution (3 mm but can be better)
one image

fMRI Blood Oxygenation Level Dependent (BOLD)
signal indirect measure of neural activity
many images (e.g., every 2 sec for 5 mins)
? neural activity ? ? blood oxygen ? ?
fMRI signal
5
Susceptibility
Adding a nonuniform object (like a person) to B0
will make the total magnetic field nonuniform
This is due to susceptibility generation of
extra magnetic fields in materials that are
immersed in an external field For large scale
(10 cm) inhomogeneities, scanner-supplied
nonuniform magnetic fields can be adjusted to
even out the ripples in B this is called
shimming

• Susceptibility Artifact
• -occurs near junctions between air and tissue
• sinuses, ear canals
• -spins become dephased so quickly (quick T2), no
  signal can be measured

sinuses
ear canals
Susceptibility variations can also be seen around
blood vessels where deoxyhemoglobin affects T2
in nearby tissue
Source Robert Coxs web slides
6
Hemoglobin
Hemoglogin (Hgb) - four globin chains -
each globin chain contains a heme group - at
center of each heme group is an iron atom (Fe)
- each heme group can attach an oxygen atom
(O2) - oxy-Hgb (four O2) is diamagnetic ? no
?B effects - deoxy-Hgb is paramagnetic ? if
deoxy-Hgb ? ? local ?B ?
Source http//wsrv.clas.virginia.edu/rjh9u/hemog
lob.html, Jorge Jovicich
7
BOLD signal
Blood Oxygen Level Dependent signal

• neural activity ? ? blood flow ? ? oxyhemoglobin
  ? ? T2 ? ? MR signal

Mxy Signal
Mo sin?
T2 task
T2 control
Stask
?S
Scontrol
time
TEoptimum
Source fMRIB Brief Introduction to fMRI
Source Jorge Jovicich
8
BOLD signal
Source Doug Nolls primer
9
Hemodynamic Response Function
signal change (point baseline)/baseline usu
ally 0.5-3 initial dip -more focal and
potentially a better measure -somewhat elusive so
far, not everyone can find it
time to rise signal begins to rise soon after
stimulus begins time to peak signal peaks 4-6
sec after stimulus begins post stimulus
undershoot signal suppressed after stimulation
ends
10
fMRI Activation
Flickering Checkerboard OFF (60 s) - ON (60 s)
-OFF (60 s) - ON (60 s) - OFF (60 s)
Brain Activity
Source Kwong et al., 1992
Time ?
11
fMRI Setup
12
fMRI Experiment Stages Anatomicals

• 4) Take anatomical (T1) images
• high-resolution images (e.g., 1x1x2.5 mm)
• 3D data 3 spatial dimensions, sampled at one
  point in time
• 64 anatomical slices takes 5 minutes

13
fMRI Experiment Stages Functionals

• 5) Take functional (T2) images
• images are indirectly related to neural activity
• usually low resolution images (3x3x5 mm)
• all slices at one time a volume (sometimes
  also called an image)
• sample many volumes (time points) (e.g., 1
  volume every 2 seconds for 150 volumes 300 sec
  5 minutes)
• 4D data 3 spatial, 1 temporal

14
Activation Statistics
Functional images
Time
15
Statistical Maps Time Courses
16
Planning an fMRI experiment
17
Thought Experiments

• What do you hope to find? 
• What would that tell you about the cognitive
  process involved? 
• Would it add anything to what is already known
  from other techniques? 
• Could the same question be asked more easily
  cheaply with other techniques?
• Would fMRI add enough to justify the immense
  expense and effort? 
• What would be the alternative outcomes (and/or
  null hypothesis)? 
• Or is there not really any plausible alternative
  (in which case the experiment may not be worth
  doing)? 
• If the alternative outcome occurred, would the
  study still be interesting? 
• If the alternative outcome is not interesting,
  is the hoped-for outcome likely enough to justify
  the attempt? 
• What would the headline be if it worked?
• What are the possible confounds?
• Can you control for those confounds?
• Has the experiment already been done? 

18
Subtraction Logic Brain Imaging

• Example simple MT localizer
• T1 View stationary rings
• T2 View moving rings
• T2 T1 motion areas
• Possible factors added
• motion
• attentional salience
• Possible factors removed
• retinal adaptation
• You must always consider the possible components
  you could be adding or affecting.
• More sophisticated designs (e.g., parametric
  designs, conjunction designs) may better address
  true contribution of components.

19
Dealing with Attentional Confounds
fMRI data seem highly susceptible to the amount
of attention drawn to the stimulus or devoted to
the task. (Well discuss possible reasons for
this in the last class)
How can you ensure that activation is not simply
due to an attentional confound?
Add an attentional requirement to all stimuli or
tasks.

• Other common confounds that reviewers love to
  hate
• eye movements
• motor movements

20
Change only one thing between conditions

• In functional imaging studies, two paired
  conditions should differ by the
  inclusion/exclusion of a single mental process
• How do we control the mental operations that
  subjects carry out in the scanner?
• Manipulate the stimulus
• works best for automatic mental processes
• Manipulate the task
• works best for controlled mental processes
• Dont do both at once.

Source Nancy Kanwisher
21
Experimental Designs
22
Block Design Sequences
Consider the simplest case, a block design with
two conditions e.g., motion localizer stationary
rings vs. moving rings lets assume 2 sec/volume

• How long should a run be?
• Short enough that the subject can remain
  comfortable without moving or swallowing.
• Long enough that youre not wasting a lot of
  time restarting the scanner.

23
Block Design Sequences
How fast should the conditions cycle?

• Every 4 sec (2 images)
• signal amplitude is weakened by HRF
• not to far from range of breathing frequency
  (every 4-10 sec) ? could lead to respiratory
  artifacts
• if design is a task manipulation, subject is
  constantly changing tasks, gets confused

• Every 96 sec (48 images)
• more noise at low frequencies
• linear trend confound
• subject will get bored
• very few repetitions hard to do eyeball test
  of significance

24
Block Design Sequences

• Every 16 sec (8 images)
• allows enough time for signal to oscillate fully
• not near artifact frequencies
• enough repetitions to see cycles by eye
• a reasonable time for subjects to keep doing the
  same thing

25
But I have 4 conditions to compare
Adding conditions makes things way more
complicated. Theres no right answer, but like
everything else in fMRI, there are various
tradeoffs. Lets consider the case of four
conditions plus a baseline.
1. Main condition epochs all separated by
baseline epochs Pro simplest case if you want to
use event-related averaging to view time courses
and dont want to have to worry about baseline
issues Con spends a lot of your n on baseline
measures
C. Random order in each run Pro order effects
should average out Con pain to make various
protocols, no possibility to average all data
into one time course, many frequencies involved
26
2. Clustered conditions with infrequent
baselines.

• sets of four main condition epochs separated by
  baseline epochs
• each main condition appears at each location in
  sequence of four
• two counterbalanced orders (1st half of first
  order same as 2nd half of second order and vice
  versa) can even rearrange data from 2nd order
  to allow averaging with 1st order

Pro spends most of your n on key conditions,
provides more repetitions Con not great for
event-related averaging because orders are not
balanced (e.g., in top order, blue is preceded by
the baseline 1X, by green 2X, by yellow 1X and by
pink 0X.
As you can imagine, the more conditions you try
to shove in a run, the thornier ordering issues
are and the fewer n you have for each condition.
27
Blocked vs. Event-related
Source Buckner 1998
28
Spaced Mixed Trial Constant ITI
Bandettini et al. (2000) What is the optimal
trial spacing (duration intertrial interval,
ITI) for a Spaced Mixed Trial design with
constant stimulus duration?
2 s stim vary ISI
Block
Source Bandettini et al., 2000
29
Optimal Constant ITI
Brief (lt 2 sec) stimuli optimal trial spacing
12 sec For longer stimuli optimal trial spacing
8 2stimulus duration Effective loss in
power of event related design -35 i.e., for 6
minutes of block design, run 9 min ER design
Source Bandettini et al., 2000
30
Fixed vs. Random Intervals
If trials are jittered, ? ITI ? ?power
Source Burock et al., 1998
31
Optimal Rapid ITI
Rapid Mixed Trial Designs Short ITIs (2 sec) are
best
Source Dale Buckner, 1997
32
Analysis of Single Trial Types

• 1) Use GLM with predictors, just as in block
  design. Compare beta weights of predictors.
• For an ROI for each trial type, compute averaged
  time courses synced to trial onset then subtract
  differences

3) Selective averaging Dale Buckner, 1997
compute mean and variance of fMRI time course
data for each trial type. Use stats to determine
whether time course is significant (different
from zero or from another condition) based on
ANOVA (no HRF assumption needed) or covariance
with HRF
It is important to randomize or counterbalance
order (i.e., ensure that each trial type is
preceded and followed by each trial type equally
often). If you cant do this, you need to
correct for overlap with more sophisticated
methods.
33
Advantages of Event-Related

• Flexibility and randomization
• eliminate predictability of block designs
• avoid practice effects
• Post hoc sorting
• (e.g., correct vs. incorrect, aware vs. unaware,
  remembered vs. forgotten items, fast vs. slow
  RTs)
• Can look at novelty and priming
• Rare or unpredictable events can be measured
• e.g., P300
• Can look at temporal dynamics of response
• Dissociation of motion artifacts from activation
• Dissociate components of delay tasks
• Mental chronometry

Source Buckner Braver, 1999
34
Summary Experiment Designs

• BLOCK
• More sensitive to amplitude of BOLD response
• More susceptible to head motion
• Wont work with all tasks (cant have oddballs,
  for example)
• EVENT-RELATED
• More sensitive to the shape of the BOLD response
• Stimulus/BOLD relationship only approximately
  linear
• Use rapid trial design with jittered timing for
  best results

35
Interpreting Your Results
36
Interpreting Results
An fMRI activation map is ALWAYS a difference
between two task or stimulus conditions a rest
or visual fixation baseline gives the most
activation, but cognitive activity during that
time is not well defined, especially in block
designs
Make comparisons of interest directly, rather
than side-by-side presentation of activation maps
37
Credits
Slides modified from Jody Culhams fMRI for
Newbies web site http//defiant.ssc.uwo.ca/Jody_
web/fmri4newbies.htm