Transcript Basis of the BOLD signal

Basis of the BOLD signal
Ciara O’Mahony and Miriam Klein
Section 1: Basics of MRI Physics
Section 2: What does BOLD reflect?
MRI scanner
Spins (Spinning protons)
• Technique at the root of MRI and fMRI: Nuclear Magnetic Resonance
 has to do with the magnetic properties of the nucleus of atoms
• Nucleus of the hydrogen atom: a single proton
• A proton has a positive electric charge, and because it spins around
itself, it produces a small magnetic field
• Miniature bar magnet with a north and south pole
Spins align with magnetic field B0
Outside scanner
Inside scanner
M0
z
Strength of B0:
30.000 (1.5T) or 60.000 (3T) times the
earth’s magnetic field (0.0005T)
Spins precess like spinning tops: Larmor frequency
Frequency of precession is directly
proportional to the strength of the
magnetic field.
M0`
ω = γ · B0
ω - frequency of precession = Larmor frequency
y - gyromagnetic ratio (constant unique to every atom, i.e. hydrogen)
B0 - strength of magnetic field
Spins are perturbed by radiofrequency pulse
We apply an electromagnetic pulse of the correct frequency
(= radio frequency (RF) pulse with Larmor frequency)
This
(a) Perturbs the distribution of the spin up and spin down
states
(b) Aligns the phase of the spins
90 degree pulse
z
z
M0
y
Mxy
y
More about RF pulse...
What means in phase?
z
z
y
•
•
The longer and stronger the RF pulse, the more energy is absorbed, and
the more the overall (red) magnetization vector M0 flips ‘away’ from the z
axes, i.e. the larger the flip angle α
We can adjust the RF pulse such that it is exactly 90° as shown here
α
y
What we can measure: T1, T2, and T2*
When RF pulse is turned off, spins want to go back to their original state, i.e.
from
to
z
z
y
Mxy
y
What will happen?
(a) Spins go back to their preferred up/down states  T1 relaxation, slow
(b) Spins dephase  T2 and T2* relaxation, quick
M0
T1: Spins go back to up/down states
z
z
y
•
•
•
y
T1 relaxation called longitudinal
relaxation: along z-axis
Absorbed energy partly given to tissue
in the form of heat and partly
retransmitted to RF receivers
Time course of returning to equilibrium
is described by exponential function 
signal gets stronger in z-direction
M0
T1 image
T1 is unique to every tissue: Time constant T1 is defined as the point where
63% of the magnetization M has recovered alignment with B0
Slow recovery in CSF and quick in white
matter
T2: Spins dephase
z
z
y
y
•
•
•
•
T2 relaxation called transverse relaxation:
in xy plane
Caused by spin-spin interactions
The loss of signal in the xy plane produces
our signal
Signal decay in xy plane described
by exponential curve
Mxy
T2 image
T2 is also unique to every tissue. Time
constant T2 is defined as the point where
63% of the magnetization in xy has
decayed.
What is T2*?
Singer et al., 2006
Mxy
The decay is faster than T2 would predict
because of inhomogeneities in the magnetic
field  what we measure is T2*
 T2* is the apparent transverse relaxation
Mo sin
T2
T2 *
time
How has all this to do with brain activity?
• If other magnetic particles are present, T2* decay is even quicker
• When a brain area is active, less magnetic particles are present
because more oxygen (oxyhemoglobin) is present (relative to
deoxyhemoglobin) and so T2* relaxation is relatively slow
• So all we measure with fMRI/BOLD from a physics point of view are
stronger or weaker inhomogeneities in the field due to more or less
oxygen being present
Mxy
Signal
Mo
sin
Take-home message part 1:
T2* task
T2* control
Stask
Scontrol
S
TEoptimum
time
•
•
BOLD is a T2*-weighted contrast
We are measuring a signal from hydrogen but
the signal we get from hydrogen atoms is
weaker when less oxygen (oxyhemoglobin) is
present
Section 1: Basics of MRI Physics
Section 2: What does BOLD reflect?
A Typical
Neuron
Where
does the
brain use
energy?
• Mostly to
restore
balance
• recycling of
transmitter
• restore ion
gradients
Atwell & Iadecola, 2002
ATP: adenosine triphosphate: mainly
produced through oxidative glucose
metabolism
How is
the
energy
supplied?
Zlokovic & Apuzzo, 1998
Capillary networks
supply glucose and oxygen
Haemoglobin
Oxyhaemoglobin: diamagnetic
Deoxyhaemoglobin: paramagnetic
What does
BOLD
measure?
Changes in magnetic properties of haemoglobin:
• more oxyhaemoglobin
increased signal
• more deoxyhaemoglobin decreased signal
SO…we are NOT measuring oxygen usage
directly
Mxy
Signal
Mo sin
T2* task
T2 *
control
Stask
S
Scontrol
TEoptimum
time
Task: relatively more oxyhaemoglobin; less field
inhomogeneity; slower T2* contrast decay; stronger
signal
Control: relatively more deoxyhaemoglobin; more field
inhomogeneity; faster T2* contrast decay; weaker signal
Haemodynamic Response
Depends On:
•cerebral blood flow
•cerebral metabolic rate of oxygen
•cerebral blood volume
Haemodynamic
Response
Function
1.‘initial dip’
2.oversupply of
oxygenated
blood
3.decrease
before return to
baseline
How is cerebral blood flow controlled?
Supply of blood is correlated with glucose and
oxygen consumption
Response is much slower than changes in
neuronal activity
Not affected by sustained hypoxia or
hypoglycemia
How is
cerebral
blood flow
controlled?
• by-products of neuronal spiking e.g. NO
• calcium signalling in astrocytes
What component of neural activity?
Local Field Potential or Spiking?
LFP: synchronized dendritic currents, averaged
over large volume of tissue
Could LFP increase without concomitant increase in
mean firing rate?
fMRI signal might reflect not only the firing rates of
the local neuronal population, but also subthreshold
activity
Overview: What are we measuring with BOLD?
the inhomogeneities
introduced into the
magnetic field of the
scanner…
 changing ratio of
oxygenated:deoxygenated
blood...
via their effect on the
rates of dephasing of
hydrogen nuclei
Where are we?
Image time-series
Realignment
Statistical parametric map (SPM)
Kernel
Design matrix
Smoothing
General linear model
Statistical
inference
Normalisation
Gaussian
field theory
p <0.05
Template
Parameter estimates
Thanks to...
• Antoinette Nicolle
• Nikolaus Weiskopf
References:
•
•
•
•
•
•
http://www.simplyphysics.com/MRI_shockwave.html
http://www.cardiff.ac.uk/biosi/researchsites/emric/basics.html
Previous year’s talks
Physic’s Wiki: http://cast.fil.ion.ucl.ac.uk/pmwiki/pmwiki.php/Main/HomePage
Heeger, D.J. & Ress, D. (2002) What does fMRI tell us about neuronal activity?Nature
3:142.
Atwell, D. & Iadecola, C. (2002) The neural basis of functional brain imaging signals.
Trends in Neurosciences 25(12):621.