BOLD fMRI
BIAC Graduate fMRI Course
October 1, 2003
Why do we need to know
physics/physiology of fMRI?
• To understand the implications of our results
– Interpreting activation extent, timing, etc.
– Determining the strength of our conclusions
– Exploring new and unexpected findings
• To understand limitations of our method
– Choosing appropriate experimental design
– Combining information across techniques to overcome limitations
• To take advantage of new developments
– Evaluating others’ approaches to problems
– Employing new pulse sequences or protocols
Developments allowing
functional MRI
• Echoplanar imaging methods
– Proposed by Mansfield in 1977
• Ready availability of high-field scanners
– Technological developments
– Clinical applicability  insurance
reimbursement  clinical prevalence
• Discovery of BOLD contrast mechanism
Contrast Agents
• Defined: Substances that alter magnetic
susceptibility of tissue or blood, leading to
changes in MR signal
– Affects local magnetic homogeneity: decrease in T2*
• Two types
– Exogenous: Externally applied, non-biological
compounds (e.g., Gd-DTPA)
– Endogenous: Internally generated biological
compound (e.g., dHb)
External Contrast Agents
• Most common are Gadolinium-based compounds
introduced into bloodstream
– Very large magnetic moments
• Create field gradients within/around vessels
– What type of contrast would this generate?
• Large signal changes: 30-50%
– Delay in activation change until agent bolus passes through MR
imaging volume
– Width of activation change depends on delivery of bolus and
vascular filtering
– Degree of signal change depends on total blood volume of area
Belliveau et al., 1990
Slice Location
NMR intensity change (CBV)
CBV Maps (+24%)
Common Contrast Agents
Compound
Longitudinal
Relaxivity
Transverse
Relaxivity
Magnetic
Susceptibility
GdCl2
1
1
1
MnCl2
0.96
3.83
0.51
GdDTPA
0.52
0.5
1
DyDTPA
0.03
0.04
1.78
1.6
-
-
0.41
0.63
40.7
4.4
15.5
148
GDTPA albumin
Iron oxide particle (3nm)
Iron oxide particle 253nm)
Hemodynamic Measures
of Blood Flow
Does blood flow serve function?
“These facts seem to us to indicate the existence of an automatic
mechanism by which the blood supply of any part of the
cerebral tissue is varied in accordance with the activity of the
chemical changes which underlie the functional action of that
part. Bearing in mind that strong evidence exists of
localisation of function in the brain, we are of opinion that an
automatic mechanism, of the kind just referred to, is well fitted
to provide for a local variation of the blood supply in
accordance with local variations of the functional activity.”
- Roy and Sherrington (1890)
Apparatus of Angelo Mosso, as described by
James (1890)
PET Imaging
PET Studies of Brain Function
• Injection of radioactive
tracer
– Measure presence of tracer
during performance of task
– Quantization of activity during
single conditions
• Advantages
– Conceptually simple
– Allows functional
measurement
• Disadvantages
Image from M. Raichle
– Invasive contrast mechanism
– Limited repeatability
– Very low temporal resolution
History of BOLD fMRI
Decreasing Relaxation Time
Blood Deoxygenation affects T2
Recovery
T2
T1
Increasing Blood Oxygenation
Thulborn et al., 1982
Ogawa et al., 1990a
• Subjects: 1) Mice and Rats, 2) Test tubes
• Equipment: High-field MR (7+ T)
• Results 1:
– Contrast on gradient-echo images influenced by
proportion of oxygen in breathing gas
– Increasing oxygen content  reduced contrast
– No vascular contrast seen on spin-echo images
• Results 2:
– Oxygenated blood in tube leads to little signal
change, either on spin- or gradient-echo images
– Deoxygenated blood leads to large susceptibility
effects on gradient-echo images
Ogawa et al., 1990b
100% O2
Under anesthesia, rats breathing
pure oxygen have some BOLD
contrast (black lines).
Breathing a mix including CO2 results
in increased blood flow, in turn
increasing blood oxygenation.
There is no increased metabolic load
(no task).
90% O2, 10% CO2
Therefore, BOLD contrast is reduced.
Spin Echo
Gradient Echo
Ogawa 1990b
Kwong et al., 1992
 VISUAL 
 MOTOR 
Ogawa et al., 1992
• High-field (4T) in humans
• Patterned visual
stimulation at 10 Hz
• Gradient-echo (GRE)
pulse sequence used
– Surface coil recorded
• Significant image
intensity changes in
visual cortex
• Image signal intensity
changed with TE change
– What form of contrast?
Blamire et al., 1992
This was the first event-related fMRI
study. It used both blocks and pulses of
visual stimulation.
Gray
Matter
Hemodynamic
response to long
stimulus durations.
Hemodynamic response to short
stimulus durations.
White
matter
Outside Head
BOLD Endogenous Contrast
• Blood Oxyenation Level Dependent Contrast
– Deoxyhemoglobin is paramagnetic, oxyhemoglobin is
less so.
– Magnetic susceptibility of blood increases linearly with
increasing oxygenation
• Oxygen is extracted during passage through
capillary bed
– Arteries are fully oxygenated
– Venous (and capillary) blood has increased proportion
of deoxyhemoglobin
– Difference between oxy and deoxy states is greater
for veins  BOLD sensitive to venous changes
from Mosley & Glover, 1995
T2* and T2 Contrast
MR
Signal
MR
Signal
T2 Decay
T1 Recovery
50 ms
1s
Discarded and Deleted Acquisitions
(DISDAQs)
Raw MR Signal (arbitrary units)
1300
1200
1100
1000
900
800
700
0
10
20
Images (TRs)
30
40
Relation of BOLD Activity to
Neuronal Activity
1. Assumptions
• fMRI response varies with pooled neuronal
activity in a brain region
– Behavior/cognitive ability determined by
pooled activity
• Alternatively, if single neurons governed
behavior, fMRI activation may be
epiphenomenal
BOLD response reflects pooled
local field potential activity
(Logothetis et al, 2001)
fMRI Hemodynamic
Response
7
100ms
6
500ms
1500ms
1500ms
500ms
100ms
5
4
3
Calcarine
Sulci
2
1
0
-6
-4
-2
0
2
4
6
8
10
12
-1
7
100ms
6
500ms
1500ms
5
4
Fusiform
Gyri
3
2
1
0
-6
-4
-2
0
-1
2
4
6
8
10
12
100
Calcarine
80
60
40
20
0
-200
0
200
400
600
800
1000
1200
1400
1600
1800
1600
1800
-20
-40
-60
-80
1500ms
500ms
100ms
-100
100
Fusiform
80
60
100
500
1500
40
20
0
-200
0
-20
*
-40
-60
-80
-100
200
400
600
800
1000
1200
1400
2. Measuring Deoxyhemoglobin
• fMRI measurements are of amount of
deoxyhemoglobin per voxel
• We assume that relative oxygenation
changes with amount of deoxygenated
hemoglobin
3. Different changes in CBF & CMRO2
• Cerebral Blood Flow (CBF) and Cerebral
Metabolic Rate of Oxygen (CMRO2) are coupled
under baseline conditions
– PET measures CBF well, CMRO2 poorly
– fMRI measures CMRO2 well, CBF poorly
• CBF about .5 ml/g/min under baseline conditions
– Increases to max of about .7-.8 ml/g/min under
activation conditions
• CMRO2 only increases slightly with activation
– Note: A large CBF change may be needed to support
a small change in CMRO2
The Hemodynamic Response
Impulse-Response Systems
• Impulse: single event that evokes changes
in a system
– Assumed to be of infinitely short duration
• Response: Resulting change in system
Impulses
Convolution
Response
=
Output
Basic Form of Hemodynamic Response
Peak
Rise
Initial Dip
Baseline
-10
-5
Undershoot
0
5
Peak
10
15
20
25
Sustained
Response
Rise
Initial Dip
Undershoot
Baseline
-10
-5
0
5
10
15
20
25
Baseline Period
• Why include a baseline period in epoch?
– Corrects for scanner drift across time
Initial Dip (Hypo-oxic Phase)
•
Transient increase in oxygen consumption, before change in blood flow
– Menon et al., 1995; Hu, et al., 1997
•
Shown by optical imaging studies
– Malonek & Grinvald, 1996
•
Smaller amplitude than main BOLD signal
– 10% of peak amplitude (e.g., 0.1% signal change)
•
Potentially more spatially specific
– Oxygen utilization may be more closely associated with neuronal activity than
perfusion response
-10
-10
-5
-5
0
5
10
15
0
20
25
Early Evidence for the Initial Dip
A
C
B
Menon et al, 1995
Rise (Hyperoxic Phase)
• Results from vasodilation of arterioles, resulting
in a large increase in cerebral blood flow
• Inflection point can be used to index onset of
processing
-10
-5
0
5
10
15
20
25
Peak – Overshoot
• Over-compensatory response
– More pronounced in BOLD signal measures
than flow measures
• Overshoot found in blocked designs with
extended intervals
– Signal saturates after ~10s of stimulation
-10
-5
0
5
10
15
20
25
-10
-5
0
5
10
15
20
25
Sustained Response
• Blocked design analyses rest upon
presence of sustained response
– Comparison of sustained activity vs. baseline
– Statistically simple, powerful
• Problems
– Difficulty in identifying magnitude of activation
– Little ability to describe form of hemodynamic
response
– May require detrending of raw time course
Undershoot
• Cerebral blood flow more locked to stimuli
than cerebral blood volume
– Increased blood volume with baseline flow
leads to decrease in MR signal
• More frequently observed for longerduration stimuli (>10s)
– Short duration stimuli may not evidence
– May remain for 10s of seconds
Issues in HDR Analysis
• Delay in the HDR
– Hemodynamic activity lags neuronal activity
• Amplitude of the HDR
• Variability in the HDR
• HDR as a relative measure
The Hemodynamic Response
Lags Neural Activity
Experimental
Design
Convolving
HDR
Time-shifted
Epochs
Introduction
of Gaps
Percent Signal Change
505
1%
500
5
5
0
5
10
15
10
15
205
• Peak / mean(baseline)
• Often used as a basic
measure of “amount of
processing”
• Amplitude variable
across subjects, age
groups, etc.
20
25
20
25
1%
200
0
5
Amplitude of the HDR
• Peak signal change dependent on:
– Brain region
– Task parameters
– Voxel size
– Field Strength
Kwong et al, 1992
Variability in the Hemodynamic
Response
•
•
•
•
Across Subjects
Across Sessions in a Single Subject
Across Brain Regions
Across Stimuli
Relative vs. Absolute Measures
• fMRI provides relative change over time
– Signal measured in “arbitrary MR units”
– Percent signal change over baseline
• PET provides absolute signal
– Measures biological quantity in real units
•
•
•
•
CBF: cerebral blood flow
CMRGlc: Cerebral Metabolic Rate of Glucose
CMRO2: Cerebral Metabolic Rate of Oxygen
CBV: Cerebral Blood Volume
Detection vs. Estimation
• Detection power
– The ability of a design to determine whether or not an
area is active
• Power: ability to detect an effect that is there (1-ß)
• Alpha: odds that a detected effect is due to chance
• Estimation efficiency
– The ability of a design to characterize the form of the
BOLD changes
– With a priori knowledge of HDR timing, shape, etc.,
one can describe temporal properties of brain regions
We can think of these challenges as spatial and temporal, respectively.
Analyzing data using BIAC
tools
Basic Steps of Analysis
• Reconstruction of K-Space Data
• Pre-processing Steps
– Slice Timing Correction
– Motion Correction
– Coregistration, Normalization, Smoothing
• Epoch averaging / Regression
– Statistical comparison of data with convolved
hemodynamic response