Introduction to fMRI

advertisement
1. Introduction to fMRI
2. Basic fMRI Physics
3. Data Analysis
4. Localisation
5. Cortical Anatomy
1
1. Introduction to fMRI
2
MRI vs. fMRI
MRI studies brain anatomy.
Functional MRI (fMRI)
studies brain function.
3
Brain Imaging: Anatomy
CAT
Brain Imaging: Anatomy
Photography
PET
MRI
4
Source: modified from Posner & Raichle, Images of Mind
MRI vs. fMRI
high resolution
(1 mm)
MRI
fMRI
low resolution
(~3 mm but can be better)
one image
fMRI
Blood Oxygenation Level Dependent (BOLD) signal
indirect measure of neural activity: active
neurons shed oxygen and become more
magnetic increasing the fMRI signal
 neural activity
many images
(e.g., every 2 sec for 5 mins)
  blood oxygen   fMRI signal5
fMRI Activation
Flickering Checkerboard
OFF (60 s) - ON (60 s) -OFF (60 s) - ON (60 s) - OFF (60 s)
Brain
Activity
6
Time 
Source: Kwong et al., 1992
PET and fMRI Activation
7
Source: Posner & Raichle, Images of Mind
fMRI Setup
8
fMRI Experiment Stages: Prep
1) Prepare subject
•
Consent form
•
Safety screening
•
Instructions
2) Shimming
•
putting body in magnetic field makes it non-uniform
•
adjust 3 orthogonal weak magnets to make magnetic field as
homogenous as possible
3) Sagittals
Note: That’s one g, two t’s
Take images along the midline to use to plan slices
9
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
10
Slice Terminology
VOXEL
(Volumetric Pixel)
Slice Thickness
e.g., 6 mm
In-plane resolution
e.g., 192 mm / 64
= 3 mm
3 mm
6 mm
SAGITTAL SLICE
IN-PLANE SLICE
3 mm
Number of Slices
e.g., 10
Matrix Size
e.g., 64 x 64
Field of View (FOV)
e.g., 19.2 cm
11
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
…
first volume
(2 sec to acquire)
12
Activation Statistics
Functional images
~2s
ROI Time
Course
fMRI
Signal
(% change)
Time
Condition
Statistical Map
superimposed on
anatomical MRI image
Time
Region of interest (ROI)
~ 5 min
13
Statistical Maps & Time Courses
Use stat maps to pick regions
Then extract the time course
14
2D  3D
15
Design Jargon: Runs
session: all of the scans collected from one subject in one day
run (or scan): one continuous period of fMRI scanning (~5-7 min)
experiment: a set of conditions you want to compare to each other
condition: one set of stimuli or one task
Note: Terminology can vary from one fMRI site
to another (e.g., some places use “scan” to refer
to what we’ve called a volume).
4 stimulus conditions
+ 1 baseline condition (fixation)
A session consists of one or more experiments.
Each experiment consists of several (e.g., 1-8) runs
More runs/expt are needed when signal:noise is low or the effect is weak.
Thus each session consists of numerous (e.g., 5-20) runs (e.g., 0.5 – 3
hours)
16
Design Jargon: Paradigm
paradigm (or protocol): the set of conditions and their order used in a
particular run
run
epoch: one instance of a
condition
first “objects right” epoch
second “objects right” epoch
volume #1
(time = 0)
Time
epoch
8 vol x 2 sec/vol = 16 sec
volume #105
(time = 105 vol x 2 sec/vol = 210 sec = 3:30)
17
2. Basic fMRI Physics
18
Recipe for MRI
1) Put subject in big magnetic field (leave him there)
2) Transmit radio waves into subject [about 3 ms]
3) Turn off radio wave transmitter
4) Receive radio waves re-transmitted by subject
– Manipulate re-transmission with magnetic fields during this readout
interval [10-100 ms: MRI is not a snapshot]
5) Store measured radio wave data vs. time
– Now go back to 2) to get some more data
6) Process raw data to reconstruct images
7) Allow subject to leave scanner (this is optional)
19
Source: Robert Cox’s web slides
History of NMR
NMR = nuclear magnetic resonance
Felix Block and Edward Purcell
1946: atomic nuclei absorb and reemit radio frequency energy
1952: Nobel prize in physics
nuclear: properties of nuclei of atoms
magnetic: magnetic field required
resonance: interaction between
magnetic field and radio frequency
Bloch
Purcell
NMR  MRI: Why the name change?
most likely explanation:
nuclear has bad connotations
less likely but more amusing explanation:
20 NMR
subjects got nervous when fast-talking doctors suggested an
History of fMRI
MRI
-1971: MRI Tumor detection (Damadian)
-1973: Lauterbur suggests NMR could be used to form images
-1977: clinical MRI scanner patented
-1977: Mansfield proposes echo-planar imaging (EPI) to acquire images faster
fMRI
-1990: Ogawa observes BOLD effect with T2*
blood vessels became more visible as blood oxygen decreased
-1991: Belliveau observes first functional images using a contrast agent
-1992: Ogawa et al. and Kwong et al. publish first functional images using BOLD
signal
Ogawa
21
Necessary Equipment
4T magnet
RF Coil
gradient coil
(inside)
Magnet
Gradient Coil
RF Coil
22
Source: Joe Gati, photos
The Big Magnet
Very strong
1 Tesla (T) = 10,000 Gauss
Earth’s magnetic field = 0.5 Gauss
4 Tesla = 4 x 10,000  0.5 = 80,000X Earth’s magnetic field
Continuously on
Main field = B0
Robarts Research Institute 4T
x 80,000 =
Source: www.spacedaily.com
B0
23
Magnet Safety
The whopping strength of the magnet makes safety essential.
Things fly – Even big things!
Source: www.howstuffworks.com
Screen subjects carefully
Make sure you and all your students & staff are aware of hazzards
Develop stratetgies for screening yourself every time you enter the magnet
Source: http://www.simplyphysics.com/
flying_objects.html
24
Subject Safety
Anyone going near the magnet – subjects, staff and visitors – must be
thoroughly screened:
Subjects must have no metal in their bodies:
• pacemaker
• aneurysm clips
• metal implants (e.g., cochlear implants)
• interuterine devices (IUDs)
• some dental work (fillings okay)
This subject was wearing a hair band with a ~2 mm
Subjects must remove metal from their bodies
copper clamp. Left: with hair band. Right: without.
• jewellery, watch, piercings
Source: Jorge Jovicich
• coins, etc.
• wallet
• any metal that may distort the field (e.g., underwire bra)
Subjects must be given ear plugs (acoustic noise can reach 120 dB)
25
Outside magnetic field
Protons align with field
• randomly oriented
Inside magnetic field
M
• spins tend to align parallel or anti-parallel
to B0
• net magnetization (M) along B0
• spins precess with random phase
• no net magnetization in transverse plane
• only 0.0003% of protons/T align with field
longitudinal
axis
Longitudinal
magnetization
M=0
Source: Mark Cohen’s web slides
Source: Robert Cox’s web slides
transverse
plane 26
fMRI Basics – The functional magnetic resonance imaging
technique measures the amount of oxygen in the blood in small
regions of the brain. These regions are called voxels. Neural
activity uses up oxygen and the vasculature responds by
providing more highly oxygenated blood to local brain regions.
Thus a change in amount of oxygen in the blood is measured,
and this is taken as a proxy for the amount of local neural activity.
The measured signal is often called the BOLD signal (Blood
Oxygen Level Dependent). Because neural activity is not
measured directly, one needs to think about what the indirect
signal really tells us, and how it’s spatial and temporal resolution
are limited. Certainly, however,the BOLD signal tells us
something about localization of neural
activity in the brain.
27
BOLD signal
Blood Oxygen Level Dependent signal
neural activity   blood flow   oxyhemoglobin   T2*   MR signal
Mxy
Signal
Mo
sin
T2* task
T2* control
Stask
Scontrol
S
TEoptimum
Source: fMRIB Brief Introduction to fMRI
time
Source: Jorge Jovicich
28
BOLD signal
29
Source: Doug Noll’s primer
3. DATA ANALYSIS
30
Hypotheses vs. Data
Hypothesis-driven
Examples: t-tests, correlations, general linear model (GLM)
• a priori model of activation is suggested
• data is checked to see how closely it matches components of the model
• most commonly used approach
Data-driven
Independent Component Analysis (ICA)
• no prior hypotheses are necessary
• multivariate techniques determine the patterns in the data that account for the most
variance across all voxels
• can be used to validate a model (see if the math comes up with the components you
would’ve predicted)
• can be inspected to see if there are things happening in your data that you didn’t
predict
• can be used to identify confounds (e.g., head motion)
• need a way to organize the many possible components
• new and upcoming
31
Comparing the two approaches
Region of Interest (ROI) Analyses
•
•
•
•
•
•
Gives you more statistical power because you do not have to correct for
the number of comparisons
Hypothesis-driven
ROI is not smeared due to intersubject averaging
Easy to analyze and interpret
Neglects other areas which may play a fundamental role
Popular in North America
Whole Brain Analysis
•
•
•
•
•
•
Requires no prior hypotheses about areas involved
Includes entire brain
Can lose spatial resolution with intersubject averaging
Can produce meaningless “laundry lists of areas” that are difficult to
interpret
Depends highly on statistics and threshold selected
Popular in Europe
NOTE: Though different experimenters tend to prefer one method over the other, they are NOT
mutually exclusive. You can check ROIs you predicted and then check the data for other areas.
32
Source: Tootell et al., 1995
Why do we need statistics?
MR Signal intensities are arbitrary
-vary from magnet to magnet, coil to coil, within a coil (especially surface coil),
day to day, even run to run
-may also vary from area to area (some areas may be more metabolically active)
We must always have a comparison condition within the same run
We need to know whether the “eyeball tests of significance” are real.
Because we do so many comparisons, we need a way to compensate.
33
Two approaches: ROI
A. ROI approach
1.
2.
3.
Do (a) localizer run(s) to find a region (e.g., show moving rings to find MT)
Extract time course information from that region in separate independent runs
See if the trends in that region are statistically significant
Because the runs that are used to generate the area are independent from those used
to test the hypothesis, liberal statistics can be used
Example study: Tootell et al, 1995, Motion Aftereffect
Localize “motion area” MT in a run comparing
moving vs. stationary rings
Extract time courses from MT in subsequent runs while
subjects see illusory motion (motion aftereffect)
MT
34
Source: Tootell et al., 1995
4. LOCALISATION
35
BRAIN LOCALIZATION AND ANATOMY
with an emphasis on cortical areas
Why so corticocentric?
•cortex forms the bulk of the brain
•subcortical structures are hard to image (more vulnerable to motion artifacts) and resolve
with fMRI
•cortex is relevant to many cognitive processes
•neuroanatomy texts typically devote very little information to cortex
Caveats of corticocentrism:
•other structures like the cerebellum are undoubtedly very important (contrary to popular
belief it not only helps you “walk and chew gum at the same time” but also has many
cognitive functions) but unfortunately are poorly understood as yet
•need to remember there may be lots of subcortical regions we’re neglecting
36
How can we define regions?
1. Talairach coordinates
2. Anatomical localization
3. Functional localization
• Region of interest (ROI) analyses
37
Talairach Coordinate System
Individual brains are different shapes and sizes…
How can we compare or average brains?
Talairach & Tournoux, 1988
• squish or stretch brain into “shoe box”
• extract 3D coordinate (x, y, z) for each
activation focus
Note: That’s TalAIRach, not TAILarach!
Source: Brain Voyager course slides
38
Rotate brain into ACPC plane
Corpus Callosum
Fornix
Find anterior commisure (AC)
Find posterior commisure (PC)
ACPC line
= horizontal axis
Note: official Tal sez use top of
AC and bottom of PC
Pineal Body
“bent asparagus”
Source: Duvernoy, 1999
39
Deform brain into Talairach space
Mark 8 points in the brain:
• anterior commisure
• posterior commisure
• front
• back
• top
• bottom (of temporal lobe)
• left
• right
Squish or stretch brain to fit in “shoebox”
of Tal system
y<0
AC=0
y
y>0
z
y>0
Extract 3 coordinates
ACPC=0
y<0
40
x
Left is what?!!!
Neurologic (i.e. sensible) convention
• left is left, right is right
L
R
-
Note: Make sure you know what your magnet
and software are doing before publishing
left/right info!
+
x=0
Radiologic (i.e. stupid) convention
• left is right, right is left
R
L
+
x=0
Note: If you’re really unsure which side is
which, tape a vitamin E capsule to the one
side of the subject’s head. It will show up
on the anatomical image.
41
How to Talairach
For each subject:
•
•
•
Rotate the brain to the ACPC Plane (anatomical)
Deform the brain into the shoebox (anatomical)
Perform the same transformations on the functional data
For the group:
Either
a)
Average all of the functionals together and perform stats on that
b)
Perform the stats on all of the data (GLM) and superimpose the statmaps on an averaged
anatomical (or for SPM, a reference brain)
Averaged anatomical for 6 subjects
Averaged functional for 7 subjects
42
Talairach Atlas
43
Brodmann’s Areas
Brodmann (1905):
Based on cytoarchitectonics: study of
differences in cortical layers between areas
Most common delineation of cortical areas
More recent schemes subdivide
Brodmann’s areas into many smaller
regions
Monkey and human Brodmann’s areas not
necessarily homologous
44
Talairach Pros and Cons
Advantages
• widespread system
• allows averaging of fMRI data between subjects
• allows researchers to compare activation foci
• easy to use
Disadvantages
• based on the squished brain of an elderly alcoholic woman (how
representative is that?!)
• not appropriate for all brains (e.g., Japanese brains don’t fit well)
• activation foci can vary considerably – other landmarks like sulci may
be more reliable
45
Anatomical Localization
Sulci and Gyri
gray matter
(dendrites & synapses)
BANK
white matter
(axons)
FISSURE
FUNDUS
Source: Ludwig & Klingler, 1956 in Tamraz & Comair, 2000
46
Variability of Sulci
Variability of Sulci
Source: Szikla et al., 1977 in Tamraz & Comair, 2000
47
Variability of Functional Areas
Watson et al., 1995
-functional areas (e.g., MT) vary
between subjects in their Talairach
locations
-the location relative to sulci is more
consistent
Source: Watson et al. 1995
48
Cortical Surfaces
segment gray-white
matter boundary
render cortical surface
inflate cortical surface
sulci = concave = dark gray
gyri = convex = light gray
Advantages
• surfaces are topologically more accurate
• alignment across sessions and experiments allows task comparisons
49
Source: Jody Culham
Cortical Inflation Movie
Movie: unfoldorig.mpeg
http://cogsci.ucsd.edu/~sereno/unfoldorig.mpg
Source: Marty Sereno’s web page
50
Cortical Flattening
2) make cuts along
the medial surface
(Note, one cut
typically goes along
the fundus of the
calcarine sulcus
though in this
example the cut was
placed below)
1) inflate the brain
3) unfold the medial
surface so the
cortical surface lies
flat
4) correct for the
distortions so that the
true cortical distances
are preserved
51
Source: Brain Voyager Getting Started Guide
Spherical Averaging
Future directions of fMRI: Use cortical
surface mapping coordinates
Inflate the brain into a sphere
Use sulci and/or functional areas to match
subject’s data to template
Cite “latitude” & “longitude” of spherical
coordinates
Source: Fischl et al., 1999
Movie: brain2ellipse.mpeg
http://cogsci.ucsd.edu/~sereno/coord1.mpg52
Source: Marty Sereno’s web page
Spherical Averaging
53
Source: MIT HST583 online course notes
5. CORTICAL ANATOMY
54
14 Major Sulci
Main sulci are formed early in development
Fissures are really deep sulci
Typically continuous sulci
•Interhemispheric fissure
•Sylvian fissure
•Parieto-occipital fissure
•Collateral sulcus
•Central sulcus
•Calcarine Sulcus
Typically discontinuous sulci
•Superior frontal sulcus
•Inferior frontal sulcus
•Postcentral sulcus
•Intraparietal sulcus
•Superior temporal sulcus
•Inferior temporal sulcus
•Cingulate sulcus
•Precentral sulcus
Other minor sulci are much less reliable
55
Source: Ono, 1990
Interhemispheric Fissure
-hugely deep (down to corpus callosum)
-divides brain into 2 hemispheres
56
Sylvian Fissure
-hugely deep
-mostly horizontal
-insula (purple) is buried within it
-separates temporal lobe from parietal and frontal lobes
Sylvian Fissure
57
Parieto-occipital Fissure and Calcarine Sulcus
Parieto-occipital fissure (red)
-very deep
-often Y-shaped from sagittal view, X-shaped
in horizontal and coronal views
Calcarine sulcus (blue)
-contains V1
Cuneus (pink)
-visual areas on medial side above
calcarine (lower visual field)
Lingual gyrus (yellow)
-visual areas on medial side below
calcarine and above collateral sulcus
(upper visual field)
58
Collateral Sulcus
-divides lingual (yellow) and parahippocampal (green) gyri from fusiform gyrus (pink)
59
Cingulate Sulcus
-divides cingulate gyrus (turquoise) from precuneus (purple) and paracentral lobule (gold)
60
Central, Postcentral and Precentral Sulci
Central Sulcus (red)
-usually freestanding (no intersections)
-just anterior to ascending cingulate
Postcentral Sulcus (red)
-often in two parts (superior and inferior)
-often intersects with intraparietal sulcus
-marks posterior end of postcentral gyrus
(somatosensory strip, purple)
Precentral Sulcus (red)
-often in two parts (superior and inferior)
-intersects with superior frontal sulcus (Tjunction)
-marks anterior end of precentral gyrus (motor
strip, yellow)
ascending band
of the cingulate
61
Intraparietal Sulcus
-anterior end usually intersects with inferior postcentral (some texts call inferior postcentral the
ascending intraparietal sulcus)
-posterior end usually forms a T-junction with the transverse occipital sulcus (just posterior to the
parieto-occipital fissure)
-IPS divides the superior parietal lobule from the inferior parietal lobule (angular gyrus, gold, and
supramarginal gyrus, lime)
POF
62
Slice Views
inverted omega
= hand area of motor cortex
63
Superior and Inferior Temporal
Sulci
Superior Temporal Sulcus (red)
-divides superior temporal gyrus (peach) from middle temporal gyrus (lime)
Inferior Temporal Sulcus (blue)
-not usually very continuous
-divides middle temporal gyrus from inferior temporal gyrus (lavender)
64
Superior and Inferior Frontal Sulci
Superior Frontal Sulcus (red)
-divides superior frontal gyrus (mocha) from middle frontal gyrus (pink)
Inferior Frontal Sulcus (blue)
-divides middle frontal gyrus from inferior frontal gyrus (gold)
Frontal Eye fields lie at this junction
orbital gyrus (green) and frontal pole (gray) also shown
65
Medial Frontal
-superior frontal gyrus continues on medial side
-frontal pole (gray) and orbital gyrus (green) also shown
66
Download