Please enable JavaScript.
Coggle requires JavaScript to display documents.
Task 5: MRI, PET, fMRI – Brain or Vein? (CM ((2) fMRI clarify terms
…
Task 5: MRI, PET, fMRI – Brain or Vein?
(3) Menon RS, Kim S-G. (1999). Spatial and temporal limits in cognitive neuroimaging with fMRI. [read only Boxes 1+2 (pp. 208-210)]
Box1. What are we measuring?
- regional differences in tonic background activity have been observed with PET
- change in local cerebral bloodflow probably triggered by transport & recycling of NT -> rCBF accepted marker, but exact relationship between neuroal activity and rCBF is unclear
- exact relationship between rCBF and MRI is not established either
- the BOLD effect is manifested in the blueish capillary beds, venules and draining veins, which are only 60-70% saturated with oxygen at rest
- Limits to reaching arbitrarily high spatial resolution
- signal can come from draining veins (not close to origin of neuronal activity ) to capillary beds (very close to neuronal activity)
- maps at columnar level are only possible if the microvascular response is limited to active neurons and if the macrovascular response is controlled for (short stimuli, vs block designs) (need to control for macrovascular response, since it is happening later than microvascular response!)
- one probably needs to look distal to the site of inhibition to see a response in fMRI signal
- we don't know how spikes and sub-treshold activation effect the fMRI signal
(4) Goebel, R. (2014). Localization of brain activity using functional magnetic resonance imaging Sections 2.1, 2.2.
Introduction
- pretty much incomprehensible without having read the rest of the text, so why bother?
- parallel imaging techniques
2.1 Physical Princriples of fMRI
- primary magnet, shimming coil (shapes magnetic field for increased homogeneity) , gradient coil, radiofrequency coil
- head coil: sends rf pulses and electromagnetic responses
2.1.1 Spin Excitation and Signal Reception
- Hydrogen-1 atom spins are randomly oriented
- B0 orients a slightly higher proportion parallel vs antiparallel to it
- -> body is magnetized
- spinning protons wobble, or precess, about the axis of the external B0 field
- If an applied electromagnetic pulse has the same frequency as the proton’s precession frequency, then the protons get “excited” by absorbing the transmitted energy.
- parallel protons flip into antiparallel state (which is higher energy) and precess in phase
- transversal relaxation due to spin-spin interactions, inhomogeneities in static magnetic field & tissue
- free induction decay (signal)
- 180° RF Pulse
- reversal of spin
- spin echo
- spin-lattice interactions
2.1.2 Image Reconstruction
- selective excitation of a slice
gradient coil modulated magnetic field strength, and thereby precession frequency. An RF pulse with the same frequency of the precession can now selectively activate an area
, frequency encoding and
- phase encoding
h-atoms now precess with a systematic phase shift along the positions within each row.
clarify
- precession: is a change in the orientation of the rotational axis of a rotating body.
- circulatory system, arteries, arterioles, capillaries, venules, veins
2.2 Physiological Principles of fMRI
2.2.1 Neuro-Vascular Coupling
- It thus seems likely that the hemodynamic response primarily reflects the input and local processing of neuronal information rather than the output signals
- post-stimulus undershoot: In the expanded space of the dilated vessels more deoxygenated hemoglobin will accumulate
2.2.2 The BOLD Effect
- activity -> overshoot -> more homogenous field
- excited spins dephase slower in homog. field -> stronger signal
- despite this sluggish response, the latency of response onsets appear to reflect quite precisely neuronal onset times
- Assuming a linear time invariant (LTI) system, one can predict the expected time course of arbitrary long stimulation periods from the known response to a short stimulus.
2.2.3 The BOLD Hemodynamic Response
Box2. Temporal Features of the fMRI response
- onset delay (~2s) ,
- time-to-peak delay (depends on stim. properties up to for 4s stimulus lenght
- (undershoot) not consistently observed
- stabilization
- inflection point: point where steepness goes from increasing its value to decreasing its value and vice versa
Rainer Goebel Lecture
- Broca's Tam Tam patient and the holographic hypothesis
- MRI (previously called Nuclear Resonance Imaging)
- protons have a spin, they rotate and create a tiny magnetic field
- all those spins are not aligned
- precession frequency is related to strength of magnetic field
- radio frequency pulse
- precessions are phase-locked
- with increased desynchronization t (free induction decay)
- T1 and T2 relaxation
- different tissue types have different de-phasing times
- two processes that remove the signal
How MRI works (5) - Spatial encoding using a gradient
- additional magnetic field to distinguish locations: each produce different frequencies
- signal that is being picked up has mixed frequencies
- FFT give location
- a lot of protons in tissue will create much signal from that frequency (high amplitude)
T1 and T2 weighted images
- weighting makes different tissues have different intensities
noise of the fMRI scanner? PET
- why is PET still important?
- labelling of specific NTs possible
Voxel-wise statistical Data Analysis
- t-test of activation differences
Brain or Vein? BOLD FMRI: Temporal Resolution time-resolved fMRI
-
(1) Gazzaniga - CN - Chapter 3 Methods pp104-110
The Marriage of Function and Structure: Neuroimaging
- When a brain area is active, more glucose and oxygen are provided by increasing blood flow to that active region, at the expense of other parts of the brain.
Positron Emission Tomography
- radioactive material emits positron, which crashed with electron, creating two photons (gamma radiation) which are picked up a sensor
PiB: A Recent Addition to the PET Tracker Family
- PET dye that marks beta-amyloid, a protein involved in Alzheimer's pathology
Functional Magnetic Resonance Imaging
- measures the ratio of oxygenated to deoxygenated blood
- makes repeated testing possible, since no radioactivity, enabling more complete statistical analysis
Block Design Versus Event-Related Design
- time for radiation to accumulate (40s) -> must use block design
- long time -> specificity suffers
- in fMRI event-related designs possible, which makes random presentation of stimulus and control trials possible (more likely to have same attentional state)
- event-related measurement allows greate range of experimental set-ups
- block is better at detecting small differences
Limitations of PET and fMRI
- comparison of experimental and control condition often produces many differences
- the BOLD signal is primarily driven by neuronal input, rather than neuronal output
(2) Ward, J. The Imaged Brain. Chapter 4. pp. 49-59
Structural Imaging
Computerized Tomography
- x-ray absorption differs across tissue types
- mostly used in clinical settings
Magnetic Resonance Imaging
MRI Physics for Non-Physicists
- a small fraction of hydrogen atoms aligns with magnetic field
- radio frequency knocks protons perpendicular to their aligned orientation
- spin in new state produces detectable change in magnetic field
- relaxation rates differ across tissue types (T1)
Functional Imaging
Basic Physiology underpinning functional imaging
atwell & iadecola 2002Positron Emission Tomography
- a positron travel 2-3 milimetres before collision
- however averaging across participants causes a spatial resolution of 10mm
Functional Magnetic Resonance Imaging
- deoxyhaemoglobin has strong paramagnetic properties that cause a distortion in the magnetic field
- undershoot may be the result of relaxation in the venous system, causing a temporary increase in deoxyh.
Which Method is best - PET or fMRI
- fMRI: noice, sensitivity to movement (therefore speech)
- signal may be distorted if neighboring areas have different magnetic properties
- PET can trace chemicals on their way through N-system
From Image to Cognitive Theory: An Example of Cognitive Subtraction Methodology
- is a subtraction really always necessary
- yes, because some areas might show random peaks as high or higher than the area responsible for the cognition of interest
Problems with cognitive subtraction
- what is the difference between no pure insertion and interaction?
EvaluationBeyond Cognitive Subtraction
Cognitive Conjunctions and Factorial Designs
What exactly is the relationship between the two?
- tasks with cognitive conjunctions are used in factorial designs?
-
(1) The Signal
- What produces the signal?
- What is the signal?
(2) How does the method work?
- How does the signal travel?
- How is the signal recorded?
(3) What does it tell us?
- about the neural basis
- about functional aspects of psychol
-
Notes Anatomy & Magnetism
- Circulatory System
- pulmonary circulation
- sytemic circulation
- macrocirculation
to and from organs
- microcirculation
within organ and tissue
terminal arterioles & venules connect to cappillaries
- heart -> arteries arterioles -> cappillaries (<->tissue) -> venules -> veins -> heart
- ferromagnetic
- paramagnetic
- one or more electrons unpaired
- pulled into an external magnetic field
- diamagnetic
- all electrons paired
- produces its own magnetic field in the opposite direction of an external magnetic field (therefor weakly repelled)
- deoxyhaemoglobin: paramagnetic
- oxyhaemoglobin: diamagnetic
antiparallel: supported by parallel vector of opposite direction
CM
-
(3) PET
- clarify terms
- neurophysiological mechanism
- signal creation
(3.3) signal creation
- radioactive material emits positron (travel 2-3mm) , which crashed with electron, creating two photons (gamma radiation) which are picked up a sensor
(3.2) neurophysiological mechanism
- glucose
- PET can trace chemicals on their way through N-system
(3.1) clarify
- isotope, half-life of a radioactive compound, ionizing radiation
(2) fMRI
- clarify terms
- neurophysiological mechanism
- Gazzaniga, Ward, Menon, Goebel
- time course of the BOLD signal
- measurement vs neural activity
(2.3) Signal origin
- deoxygenated blood = low fMRI signal
- oxygenated blood = high fMRI signal
- If the brain is in resting state 30-40% of the oxygen is extracted from the blood in the capillary bed.
Image collection:
(2.4) time course of fMRI signal
- Q: always the same? any constant? how does it vary?
- Q: for the same stimulus, across regions, across individuals
- initial dip (explain)
- overcompensation
- peak
- undershoot: In the expanded space of the dilated vessels more deoxygenated hemoglobin will accumulate
(Ward & Goebel) (but also not consistently observed)
- stabilization
- talk about four interval in Menon B2?
(2.2) neurophysiological mechanism
- accepted marker, unclear relationship, capillary beds and draining veins
- synaptic integration better predictor than action-potentials
(2.1) clarify
- circulatory system, circulatory system, arteries, arterioles, capillaries, venules, veins
- paramagnetic, diamagnetic
(1) MRI
- clarify terms
- signal creation
- image collection
so is it only water or also fat?
Ward says it is H in water that produces the signal
(2.3) image collection
- T1 & T2 weighted images
- T1 grey m. : grey white m. : white
(2.2) signal creation
the scanner: main magnet, gradient coil, radio frequency coil: send & receive
- tissue types
- hydrogen atom & their spin
- constant magnetic field
- alignment : slightly more align parallel
- precession around axis of B0
- precession frequency rel. to strength of B0
- radio frequency impulse
- same frequency -> excitation
- relaxation
- radio frequency impulse
youtu.be/djAxjtN_7VE?t=5m20s
-
(4) PET vs fMRI
- temporal resolution - spatial resolution
- invasiveness
- experimental procedure
- limitations
- biological vs technical limit
- more difficult to create homogenous magnetic field at higher strengths
- correspondance level to neural activity
(4.4) limitations
- poor temporal resolution (comb, with EEG necessary)
- comparison of experimental and control condition often produces many differences
- since regional differences in tonic background activity have been observed? (see 3)
- the BOLD signal is primarily driven by neuronal input, rather than neuronal output (input areas might not be seen in fMRI scan)
- fMRI: noice, sensitivity to movement (therefore speech)
- signal may be distorted if neighboring areas have different magnetic properties
(4.2) invasiveness
- PET less expensive
- scanner easier to maintain
(4.3) experimental procedure
- PET
- repeated testing not possible
- time for radiation to accumulate -> must use block design
- fMRI
- repeated test possible -> more complete statistical analysis
- controlling for macrovascular response (no use of block design)
- limited by time is takes for fMRI signal to return to zero
- cognitive subtraction
- goal: isolate specific cognitive activity, by subtraction activity during two tasks: comparison to rest/ other task that does not employ specific cognitive activity, but has other components that are also active during task which has component of interest
- problems: pure insertion and interaction
(4.1) temp. res. / spat. res.
- in the order of seconds - in the order of minutes
spat. res
- 1mm dep. on size of voxel - 10mm