fmri

Complete Study Guide

Page history last edited by Nicholas Davis 1 yr ago

Chapter 1
Chapter 2
Chapters 3, 4, 5
Chapters 6 & 7
Chapter 8: Spatial and Temporal Properties of fMRI
Chapter 9 focus on following:
Chapter 10 focus on the following
Chapter 11: Experimental Design
Chapter 12: Statistical Analysis
Additional study

Chapter 1

1. Describe the goals and methods of phrenology. What concept did the phrenologists introduce?

Phrenologists tried to map brain functions in accordance with the amount of brain tissue devoted to a cognitive function that influenced behavior. They assumed that increases in brain size would translate into measurable bumps on the skull. Phrenologists, despite collapsing on scientific grounds, were important because they introduced the idea of localization. Localization is the idea that the brain may have distinct regions that support particular mental processes.

2. What is functional magnetic resonance imaging?
BACK TO TOP

fMRI is a neuroimaging technique that uses standard MRI scanners to investigate changes in brain function over time. From this, one can create images of the functional organization of the brain

3. What are the limitations of lesion studies of the brain? How can functional neuroimaging help overcome these limitations?

BACK TO TOP

A well-appreciated problem results from the network structure of the brain: The fact that damage to area X impairs behavior Y indicates that X is necessary for Y, but not that X is sufficient for Y. A given brain region may support more than one function, and each function may be supported by multiple brain regions. Another problem is finding patients with isolated brain damage. Many patients have diffused damage from strokes or trauma. In addition, a lesion's effects can diminish over time through healing or compensation by other brain regions. Trying to find the context in human lesion studies is difficult. However, by using fMRI in concert with TMS, a researcher can create temporary, localized lesions and avoid some of the above limitations.

4. What are the limitations of drug studies of the brain? How can functional neuroimaging help overcome these limitations?

BACK TO TOP

A main disadvantage of drug studies is the difficulty in identifying functions of specific brain regions following systemic application of a drug. Also, many drug manipulations have relatively slow time courses, with functional changes that can take place over weeks, so inferences about short-term cognitive processes become challenging. Using animal and other techniques to identify localization and manipulation can help centralize the identifying functions of the regions.

5. What are the limitations of electrophysiological studies of the brain? How can functional neuroimaging help overcome these limitations?

BACK TO TOP

Electrophysiological methods suffer from a trade-off between localization accuracy and invasiveness. Single-unit studies allow very precise localization of a activity to a specific cell in a specific brain region, but require the insertion of electrodes directly into the brain and are thus restricted to animal studies. While extra-cranial EEG and MEG studies do not damage the brain, it is mathematically impossible to uniquely identify the locations of the neural sources that cause a given pattern of activity on the skull (the inverse problem). Animal usage and complimentary techniques, such as combining fMRI and EEGin one study, can help overcome this limitation.

6. To what aspect of imaging does “contrast” refer? How could a single image be high-contrast in one sense and low-contrast in another?
BACK TO TOP

Contrast is the intensity difference between different quantities being measured by an imaging system. It also refers to the physical quantity being measured. Contrast is expressed with respect to variation in contrast due to noise and to discuss results in terms of the magnitude of the intensity different quantities divided by the variability in their measurements (contrast-to-noise ratio). Images are created that distinguish between active and non-active areas of the brain. Functional contrast provides information about a physiological correlate of brain function (deoxygenated blood). An image may have high contrast-to-noise despite small absolute intensity differences if there is very little variability within each property being measured.

7. What is the difference between structural contrast and functional contrast?

BACK TO TOP

In structural contrast, intensity difference is based on the physical brain structure. Whereas in functional contrast, intensity difference is based on the physiological brain function (e.g. blood oxygenation change).

8. What are voxels?

BACK TO TOP

Voxels are the basic sampling units of MRI, also known as a three-dimensional volume element.

9. What is functional resolution, and how is it different from spatial and temporal resolution?

BACK TO TOP

Functional resolution is the ability to map measured physiological variation to underlying mental processes. It differs from spatial resolution, as it extends beyond spatial locations and from temporal resolution, as it is over periods of time. However, spatial and temporal resolution, along with other properties, are included in functional resolution.

10. What is resonance?

Resonance is tendency of a system to oscillate at maximum amplitude at a certain frequency. In brain imaging, magnetic resonance is the absorption of energy from a magnetic field that oscillates at a particular frequency. Resonant frequency is the oscillation that provides maximum energy transfer to the system.

11. Why did physicists use oscillating magnetic fields to study magnetic resonance effects?

BACK TO TOP

If the frequency of the oscillating magnetic field matches the spin frequency of the atomic nucleus, then the nucleus would absorb energy from the field. The resonant frequency needed for the oscillating field depends upon the strength of the static magnetic field. Thus, the frequency of the oscillating field constant and the strength of the static field is changed by adjusting the current in the magnet.

12. Describe the experimental apparatus used by Felix Bloch and his colleagues to measure nuclear magnetic resonance effects.

BACK TO TOP

Bloch places a sample of water in a brass box between the poles of a strong magnet, whose field strength they could manipulate. An adjacent transmitter coil sent electromagnetic energy into the sample, while a second detector coil was used to measure changes in the energy absorbed by the water (as emitted back to the environment). The sample was presaturated for 24 hours in the magnetic field to ensure that relaxation would occur. Bloch labeled the effects nuclear induction, and nuclear magnetic resonance (NMR).

13. How did Damadian and Lauterbur each contribute to the development of MRI?

BACK TO TOP

Damadian hypothesized that similar differences might be observed in the water molecules between cancerous and noncancerous cells during NMR – thus, making NMR a cancer detector. It was the first clear biological application for NMR. Moreover, it created images providing information of how the quantities varied over space. Lauterbur recognized that NMR had considerable potential, if an image method could be developed. By measuring how much energy was emitted at different frequencies, one could identify how much of that object was present at each spatial location. The idea of inducting spatial gradients in the magnetic field proved to be the fundamental insight that led to the creation of MR images. He also realized that a single gradient could only provide information about one spatial dimension, thus the need for multiple gradients.

14. What is an image? What advance was most critical to the development of techniques for image formation in MRI?

BACK TO TOP

An image is a visual description of how one or more quantities vary over space. Mansfield’s echo-planar imaging (EPI) allowed collection of an entire two-dimensional image slice by changing spatial gradients rapidly following a single electromagnetic pulse from a transmitter coil. This creates a complex MR signal that can be transformed using a Fourier transform in order to create a meaningful image.

15. What were the two reasons for the change in terminology from “nuclear magnetic resonance” to “magnetic resonance imaging”?
BACK TO TOP

The term NMR was abandoned in large part due to the negative health connotations of the word nuclear, which was justified because NMR does not use ionizing radiation. Moreover, the change can be attributed to the desire of hospital officials to separate MR scanning from nuclear medicine departments.

16. Why was there a boom in MRI use in the 1980s? How did this growth set the stage for fMRI?

BACK TO TOP

MRI scanners were approved in 1985 by the FDA for clinical use. This allowed MRI scans to be prescribed by physicians and billed to insurance companies and Medicare. Rather than having to subsidize the enormous cost of a scanner that was used for research, hospitals now saw them as a source of profit. This made structural MRI one of the most common diagnostic imaging procedures. Over the next decade, thousands of MRI scanners were in hospitals. This proved invaluable to the growth of interest in fMRI.

Chapter 2

1. What are the three main components of an MRI scanner?

A) The Static Magnetic Field – magnetic field creation

B) Radiofrequency Coils – the signaling mechanism

C) Gradient Coils – the framing tool

2. How does an MRI scanner generate the main magnetic field? What two criteria are important for the main magnetic field?

The fMRI creates the main magnetic field by passing current through very tightly coiled wire (electromagnetism.) The two criteria that are important are homogeneity and strength.A homogeneous magnetic field is one that has the same strength throughout a wide region near the center of the scanner bore. If there was a non uniform magnetic field, then location in the scanner would dictate the image results.

3. Why are superconducting electromagnets necessary for MRI?

Superconducting electromagnets are necessary because with zero resistance (due to extreme cooling) a very strong electric current can be passed through. This helps keep electric cost down.

4. What is the difference between a surface coil and a volume coil?

A surface coil is a single loop that consists of a inductor (L) – capacitor (c) circuit. This allows radiowaves to be sent to and received from the target molecules. The volume coil implements the exact same mechanism except it occurs over the whole volume of the head area. This allows for greater surface coverage albeit at lower resolution due to the fact that radiofrequency cables do have resistance and therefore exclude excess energy as heat.Following from their strengths and weaknesses, surface coils would be better employed when looking at one particular area of interest, while volume coils are better equipped to look at brain activity as a whole because the images it collects are more homogeneous. The optimal set up is to use volume coils to excite and surface coils to collect.

5. Why are gradients necessary for image generation? What sorts of coils are used to generate these gradients?

A gradient is necessary in order for the MR signal to become spatially controlled. Therefore, different molecules in the bore will contribute different signals depending on where they are located on the subject. The two types of coils are Maxwell and Golay pair. The basic distinction is the direction in which the gradient is created.

In Maxwell, the gradient is created in the direction (parallel to) of the main magnetic field.

In the Golay pair, however, the gradient is created perpendicular to the main magnetic field.

6. What is shimming, and why is it important?

Shimming is the process of correcting for inhomogeneities in the main magnetic field. These shimming coils are able to create first, second, and third order fields depending on the need. Each subject needs calibration as they will distort the field in different ways.There will naturally be incongruities in the field and shims are electromagnetic coils that compensate for these flaws in the static magnetic field.

7. Why might researchers want to monitor physiological changes like cardiac and respiratory rate during an fMRI experiment?

For two reasons: to improve quality of data and to deduce other cognitive functions. Sweating, breathing, etc. can all distort the data if not compensated for. Also, physiological changes like pupil dilation can help the experimenter deduce simple things such as arousal and novelty. Usually health concerns are not considered in fMRI as unhealthy participants are filtered.

8. Describe the procedures of a typical fMRI experiment, beginning with recruitment of the subject.

A) Subject calls and is informed about the purpose of the experiment

B) An initial screening can be done just to check for aneurysm clips, tattoos, pacemaker

C) At the hospital subject is screened for metal again then put in scanner and told about emergency joystick, and a volume coil is placed over her head.

D) The first scans are simply structural, only the following ones are for the experiment.

E) After the experiment the subject can discuss the results with the conductor

9. What sorts of conditions/problems would prevent someone from being a subject in an fMRI experiment?

In short, if somebody has anything metal in or on their body they should not be allowed in the scanner. Also, if somebody shows previous, significant health problems they should probably be not allowed in the scanner as it could trigger a relapse. Some claustrophobic subjects are let go before the experiment begins. Also people that have internal ferromagnetic structures/devices, such as metal plates and aneurism clips are not used in experiments.

10. What effects do very strong static magnetic fields have upon human tissue?

Very little effects. Although innumerable studies have set out to show some ‘causation’ between static magnetic fields and health problems there have not been any solid conclusions. The only real risk involves very sudden movements of the head which might cause torque on hair in the ears and possibly the rods and cones in the retina. Some reported effects are vertigo, phosphenes, metallic taste sensations, sensitivity in teeth fillings, nausea, and headaches though these all occur when somebody move their head quickly in a static field. One cause could be the magnetohydrodynamic phenomena.

11. Forty five percent of subjects reported unusual sensations when entering the bore of a 4-T scanner. What was most important about this result?

Essentially that the scanner has a huge placebo effect. People reported ‘unusual sensations’ when in reality the magnet was down for repair.

12. What happens to metal brought within the static magnetic field? Consider both large external objects (e.g., oxygen canisters) and small internal devices (e.g., aneurysm clips).

Large objects will generally just fly towards the fMRI machine. This is known as ‘projectile effect.’ Other, small objects that simply have iron in them will rotate in order to align with the magnetic field. This process is known as torsion. Obviously, if an aneurysm clip tries to rotate out of position the patient will most likely bleed to death.

13. What effects do the changing gradient fields have upon the human body? How can these effects be minimized?

Since the human body is a conductor, rapid changes in the magnetic field have the potential to cause some tingling sensations in the body. These are due primarily to peripheral nerve or muscle stimulation. The change in magnetic field over time, dB/dT, is therefore kept at sub-threshold values to make sure patients do not have this feeling. The other primary concern is with patients who have medical devices such as pacemakers. An induced current has the potential to interfere with the voltage of such devices. Anyone with a medical device in their body should be excluded from the experiment.

Human blood is an electrically conductive fluid. When these kinds of fluids are moved through a changing magnetic field, a magnetic field is created in the opposite direction. Currents induced in the body can cause peripheral nerve or muscle stimulation. In high magnetic fields this may become painful for the subjects. However, these results can be minimized by slowly moving the subject into and out of the scanner. The subjects are also instructed not to clasp their hands or cross their legs during scanning because these actions create conductive loops that may increase the effect of the magnetic gradient.

14. What is SAR? Why is it important for fMRI?

Specific Aborption Rate. “A quantity that describes how much electromagnetic energy is absorbed by the body over time.” It is important in fMRI because when the machine directs radiofrequencies at the subject his/her body will absorb some of them and his/her body temperature will increase. Looped objects can increase absorption so clearly necklaces cannot be worn. The other way to reduce SAR involves determining the strength of both the static magnetic fields and the radiofrequencies.

15. Why is it important to avoid looping wires or necklaces near the head coil?

As stated above, loops are really good at focusing electromagnetic energy from the coil and therefore can cause ‘local burning’ on the subject. This is because wires and necklaces, even if they are not ferromagnetic, heat up more than the surrounding tissue.

16. What is the most common health consequence for MRI studies? How can it be minimized?

Claustrophobia is the biggest concern. This can be reduced by first screening any subjects who say they are claustrophobic. In order to reduce anxiety in general the experimenter should explain some of the noises to the subject in the scanner and maintain frequent communication. Also the subject should be informed that he/she can get out at anytime. If the subject feels he/she has ‘some’ control over the experiment then he/she will be that much more relaxed. Also, the experimenter could walk into the room to try to calm the subject and should always pay attention to telltale anxiety signs such as the participant asking “how much longer.” If at any time the subject declares that they are anxious, then they have to be removed from the scanner immediately. A few methods that researchers have employed to reduce anxiety includes: talking with the participants before the experiment to see if they are claustrophobic; talking with the subjects while they are in the scanner; directing air-flow in the scanner to reduce the fear of suffocation; providing a panic button for the subject.

Chapters 3, 4, 5

What is resonance (simple explanation)? What determines resonance in MR and how is it used to select slice. (Ch3, q5, 8, 10, 12)

BACK TO TOP

1) Resonance is when two systems are oscillating at the same frequency. In order to have the most efficient excitation of nuclei during an MR scan, a resonant frequency must be attained.

2) This can be done by calculating the Larmor frequency.

a. The Larmor frequency takes into account the gyromagnetic ratio of a nucleus, which is the ratio between the charge and mass of a given nuclei.

b. This equation gives the most efficient amount of magnetic radiation that will reach resonant frequency and excite low energy spins to a higher energy state. When the bombardment of energy stops, the particles in a high energy state will return back to their low energy state and emit a photon that is able to be detected.

3) Resonance is a phenomenon in physics that can be applied to many systems, but in essence deals with principles of energy transfer at particular frequencies. In MRI, the frequency of the electromagnetic pulses is the same, or resonates with, the frequency of the precession of nuclear spins in a magnetic field, which allows the application of small amounts of energy to the nuclei to result in large changes in the system overall. Resonance in MR is determined by the Larmor frequency. It is the frequency at which certain atomic nuclei or particles are excited (absorb photons) to a different energy level, and thus allows the exchange of energy with the magnetic pulse. In MRI, this nucleus is the hydrogen atom, which consists of a single proton. The Larmor frequency also describes the rate at which the proton is precessing (which happens because it is a charged particle in a magnetic field). This frequency depends upon the gyromagnetic ratio, which in turn depends upon magnetic and rotational properties (charge and mass) of the nucleus.

a. In MR, energy is transmitted via RF pulses, set at the resonant frequency, or the Larmor frequency. There are two states that the nuclear spin can be in, and when energy is absorbed at just the right frequency, the state switches into the higher-energy state. This is the excitation of the protons, which is done to only a thin slice by adjusting the frequency and magnetic field. Because only those protons receiving the RF pulses at the resonant frequency will be excited, it allows the excitation of a slice of the brain instead of the whole area at one, because other locations will experience slightly different frequencies due to the gradient.

How is location determined within slice (e.g. see ch4 q 7, 16)

1) A magnetic gradient is applied across the magnetic field to create varying changes in spins that inform the detector where they came from.

2) There need to be several gradients in both the x and y direction in order to accurately localize a signal and translate that signal into k-space. There are two distinct kinds of gradients for each direction:

a. Phase encoding gradients are put into place before data acquisition. They are responsible for introducing a variety of phases to a given number of spins in the slice. This is used to control the Gy gradient which is responsible for the y-axis intervals in k-space.

b. Frequency encoding gradients, on the other hand, are responsible for creating a spin precession gradient. This is turned on during data acquisition and contributes to the Gx gradient which, in turn, determines the x-axis of k-space.

3) Spatial information is determined from both the phase of the magnetization and the frequency of the MR signal. By using a gradient magnetic field, the phases and frequencies of protons in different locations can be localized. The two methods that can localize the signal being detected by the receiving coil are phase encoding and frequency encoding, which can be separated according to their timing during the image formation process.

4) Phase encoding is based upon the different rates of precession of spins. A stronger magnetic field will cause the precession to speed up. This gradient is applied prior to data acquisition. In contrast, frequency encoding relies upon the differential effects of a gradient magnetic field that applied during data acquisition. When the frequency encoding gradient is switched on during the data acquisition, different regions experience different strengths of magnetic fields. The amplitude of the signal that is detected depends upon how many spins in a certain location are aligned with the gradient.

5) Information from 3 dimensions, provided by 1) slice excitation, 2) frequency encoding, and 3) phase encoding, is used to create an image of spatial location based on frequencies and phases. This is encoded, in that a Fourier transform is used to create the image itself from the raw data.

Be able to explain difference between T1 vs. T2 (ch3, q14-16; ch 5, q6-10)

1) T1, T2, and T2* provide different kinds of information about the brain.

2) T1 is accurate in the structural depiction of the brain and is often used to create a reference map for a certain individual in which T2*, which takes into account BOLD responses, can be co-registered with in order to line the regions up. It is good for differentiating between tissues.

a. T1 measures the decay of the signal over time as the longitudinal magnetization returns to its equilibrium state. Different tissues have different T1 values (meaning the protons in different tissues take different amounts of time to return to equilibrium from an excited longitudinal magnetization). An intermediate TR is needed, along with a small TE to minimize contrast due to T2 and T2*.

b. T1 images require a short TE to minimize any T2 data.

c. T1 requires a delicate balance of TR because this dictates the amount of time after excitation. If the data is collected too soon, a short TR value, then neither of the tissues will have relaxed, and there will be no data. However, if the TR is too long, then both of the tissues will have completely relaxed and no structural difference will appear between them. Thus, an intermediate TR value is necessary for obtaining differences between tissues with T1.

3) T2 is used to collect information on fluid filled regions of the brain. This is often used to diagnose patients with tumors or other abnormalities. These images can also be used as a reference in fMRI studies.

a. T2-weighted images show the different times between tissues in how long it takes for transverse magnetization to return to an equilibrium state. The length of time this takes differs according to the type of tissue, and the decay of the signal at different points in time can thus give information about which type of tissue is sending the signal depending on its strength. An intermediate TE is and a long TR is needed.

b. If the TE is too short, then transverse magnetization( the x,y component of net magnetization) will be the same as during excitation, no difference will be registered, so no T2 data will be acquired. Conversely, if they TE is too long, the transverse magnetization will be lost and no data can be acquired. A middle ground must be found in order for a successful T2 image to be acquired.

c. TR has to be very long in order to not collect any T1 data.

4) T2* weighted images reflect the BOLD response in the brain. They are sensitive to the amount of deoxygenated hemoglobin present. The amount of this changes as a response to neurons firing from activity related to cognitive processes (albeit quite indirectly).

What is a fourier transform?

1) A mathematical equation capable of changing a frequency domain into a time domain and vice versa. In fMRI, it is used to convert data in the time domain to data in a frequency domain. This frequency domain is referred to as the power spectrum. Raw fMRI data show the relative intensity of the signal at each time sample. The power spectrum represents this same data, but in a frequency domain, meaning that the power spectrum indicates the intensity of the signal at each component frequency.

2) A Fourier transform is the conversion of data in the time domain (which is how it is collected), into the frequency domain, by modeling it as a sine wave. The magnetic gradients and RF pulses focused on a slice allow differentiation of locations based on the phase and the frequency data collected from each coordinate. The Fourier transform is needed to convert the raw values into phase and frequency information that can be displayed as a wave

3) Fourier transform can be used to convert MR data into a K-space, which is a frequency representation of the data.

How does k-space differ from ‘normal space’ illustrate this with some concrete examples (see ch4, q’s 5, 11, 12, 14)

1) A k-space representation does not correspond to normal space. A location in the upper right corner of k-space does not mean that data is located in the upper right corner of the brain.

2) K-space is a frequency representation of MR data signals transformed by the fourier transform. It is used to create a 2D image of the data once it is analyzed.

3) The biggest difference is that the location of coordinates in k-space does not correspond in a straightforward manner to that of ‘normal’ space. In short, k-space is a graph of spatial frequency. That is, k-space is a representation of normal space that cannot be visualized as a translation or any simple functional operation. Although a k-space image may be shown as, for instance, a 3x3 inch 2D diagram that is transformed into a 3x3 inch image, a coordinate on the k-space graph does not determine the coordinate in that same location in Cartesian space on the image.

4) Each coordinate of k-space represents the summation of the gradient fields’ effects on the MR signal for all of the voxels at that location. Data is transformed from k-space to normal space via a Fourier transform, which depends on phase and frequency data. Large values in k-space are found in the center, and the smaller the value the further from the center the coordinate will be located. The larger the dimensions to the k-space image (the larger the amount of information in the periphery), the more detailed the resulting image will appear. In short, the intensity of the signal determines the central part of k-space, and the level of detail determines the periphery.

Why are slices interleaved, how does this affect temporal characteristics of fMRI? (for first part, see Ch4, q15)

1) Slices are interleaved because surrounding regions are often partially excited. If one then wants to gather information from the next slice and it is re-excited, the data acquired will be skewed because there was not enough time for the molecules to relax and come back to a baseline condition.

2) This affects the temporal characteristics of fMRI because two regions that are right next to each other may be collected seconds apart from each other. This means that by the time data is collected form a certain region, the activation may not be nearly as high as it was in the initial process when its neighbor was collected. This can create problems when trying to infer cognitive processes from fMRI experiments.

3) Slices are interleaved in order to prevent surrounding areas outside of a slice becoming saturated with MR signal, thus eliminating meaningful data from being collected. It does increase the time needed, as samples are not as spatially close to one another and it takes time to change locations back and forth.

What is TR and TE (Ch5, q3,4)

1) TR means repetition time. This is the time interval between successive excitatory pulses.

2) TE means echo time. This is the time interval between the excitatory pulse and receiving the data.

What is a susceptibility artifact? (Ch5, q18)

1) MSAs are loses in the MR image due to field imhomogeneity resulting from the boundary between brain tissue and air-filled cavities. For example, signal loss in the ventral frontal region is due to the sinus cavities.

Chapters 6 & 7

Focus on location in the brain, and ability to actually recognize and label views of the brain (q’s 9-12, but make sure you know what these look like!)

1) Front of the brain- Anterior/ rostral

2) Back of the brain - Posterior/ caudal

3) Top of the brain- Superior/ dorsal,

4) Bottom of the brain -Inferior/ ventral,

5) Close to the midline- Medial

6) Close to the sides- Lateral

Sagital: A slice from caudal to rostral

Coronal: from dorsal to ventral orthogonal to sagital

Axial: Horizontal plane

Occipital Lobe - The posterior lobe of the brain that deals with visual processing

Temporal Lobe - The lobe on the ventral surface of the cerebellum that deals with auditory and viual processing, language, memory, and other functions

Parietal Lobe - The posterior-dorsal lobe that is important for spatial processing, cognitive processing, and many other functions.

Frontal Lobe - The most anterior lobe of the cerebellum, which deals with executive processing, motor control, memory, and other functions

Limbic Lobe- Related to emotional processing and olfactory.

*Provide a coherent explanation of the steps relating neuronal activity relates to BOLD signal. Include the following:

What influences blood flow? (Ch6, q18)

Neuronal activation causes the surrounding arterioles to dilate. This dilation can be as large as a 33% increase that was found in rats by Ngai, Winn, and colleagues. The authors had a sensory stimuli and measured the arteriole dilation increases as a function of neuronal activity. They found that not only did the local arterioles dilate and increase blood velocity, but arterioles upstream from the region they were inspecting dilated as well. This is bad news for the spatial resolution of MRI when using hemodynamic blood response as an indicator of the spatial indicator of neural activity. This experiment showed that a relatively large region around an active neuron has increased blood flow, therefore limiting the spatial resolution that can be achieved using an MRI scanner.

What magnetic properties is the BOLD sensitive to? (Ch7, q1)

Oxygenated hemoglobin is diamagnetic which means it has no unpaired elections and zero magnetic moment; deoxygenated hemoglobin is paramagnetic which means it has both unpaired electrons and a significant magnetic moment. Bold is sensitive to deoxygenated hemoglobin.

What does BOLD stand for? (Ch7, q 6)

Blood-oxygenation-level dependent

What causes BOLD contrast? (Ch7, q7)

BOLD contrast is due to the difference in signal on T2 weighted images as a function of the amount of deoxygenated hemoglobin. Neuronal activity causes increased metabolic demands and thus, increased oxygen consumption. This increases the amount of deoxygenated hemoglobin, given a constant blood flow. The second mechanism is that increased blood flow in the absence of increased metabolic demand would decrease the amount of deoxygenated hemoglobin. Thus, the difference we see is a qualitative measure of the amount of deoxygenated hemoglobin.

What influences spatial resolution of BOLD? (Ch7, q9)

Oxygenated blood is distributed to more of the brain than just the active parts. As was seen in the previous study, arteries that were upstream of the activation also dilated, proving thtat the spatial resolution of BOLD cannot be fully trusted.

Why might BOLD be thought of as a quirk? (Ch7, q 12)

There must be an uncoupling of oxygen supply and oxygen consumption for BOLD contrast to be useful for functional neuroimaging because fMRI is based on the detection at the macroscopic level of changes in the microscopic magnetic fields surrounding red blood cells. Going from oxygenated to deoxygenated changes its magnetic properties and thus, is the reason we can see contrast. Measuring the BOLD response is a very indirect thing. We are really measuring a chain of reactions of which we witness the end product and hope to infer with a certain degree of accuracy, for better or worse, the abstract causes.

What special type of BOLD effect might have better spatial resolution, but has yet to be clearly seen (Ch7, q13)

Initial dip = initial increase in deoxyhemoglobin. The initial dip is hard to demonstrate conclusively because it isn’t frequently observed as it is rare to have high-field (>4T) MR scanners for functional studies.

Chapter 8: Spatial and Temporal Properties of fMRI

1) What are the major divisions of fMRI data, from subjects down to voxels?

-Spatial vs. temporal information: Finding a location of activation is different than looking at how that activation changes over time. The hemodynamic response is an example of how an activation changes over time. There is an average hemodynamic curve that roughly corresponds to the increase blood flow and gradual drop off of that increased blood flow. However, if the subject is doing multiple things, or there are different brain regions activated, there might be different components adding to the hemodynamic response and image analysis sets out to try to sort out these different acticavations over a temporal domain.

-Functional vs. structural information:MRI scans are used to gather structural information about a persons unique brain in order to analzye the regions of interest. However, if one want to know about the function of certain regions over time, one must do a functional analysis consisting of a time series of images. No one functional image can give data about the changing brain state of a subject; multiple images and activations need to be compared in order to deduce meaningful informaiton.

-Many subjects for more complex studies and less for simpler studies: in order to try to generalize over a population, one must have a sufficient average to make any sweeping claim that "x region is responsible for y cognitive activation"

-Single sessions or multiple sessions: usually, a participant will undergo one session, but if the study is looking at long term effect, like with drug effects or memory reserach, a single subject may be required to do multiple fMRI session in order to gather enough data to come to a conclusion.

-less slices for studying specific regions or more for studying the whole brain.

-Block designs vs.event-related designs: Block design presents a stimulus for a certain period of time and then rest and repeat either the same or different stimuli. Event related designs show a stimuli and observe the changing hemodynamic response it provokes, stimuli are presented in succession so that the hemodynamic response does not return to baseline and the tasks sum to change the overall hemodynamic response.

-Voxelwise analyses vs. region-of-interest (ROI) analyses: voxelwise analysis looks at the overall activity of the brain in order to get a idea of where the activation is taking place, whereas ROI analsysis already has a hypothesis as to where an activation is taking place and the researches analyze certain regions to look at activation to empirically validate their hypothesis.

2. What are disdaqs, and why are they sometimes incorprated into imaging protocols?

“Discarded data acquisitions” This concept refers to discarding the images collected at the very beginning of an fMRI study. The MR signal is the strongest at the point and the net magnetization following the excitation period has not reached a steady state, so the data is not realiable.

3. What are the disadvantages of high spatial resolution in fMRI?

Reduced signal to noise ratio and increased acquisition time. The more spatially specific an fMRI scan is, the longer it takes to acquire the images. The converse is true for temporal resolution. Better temporal resolution decreases spatial resolution.

4. What are partial volume effects?

The combination of signal contribution from two or more tissue types within one voxel. When measuring a particular voxel, one measures the entirity of that voxel, even the smallest voxels may contain two different types of tissue. In fMRI we strive to differentiate between different tissue types (T1 scan) and the distinction between fluid (T2) in order to set the groundwork for interpretting the functional image (T2*). Also, there may be brain tissue wihtin a voxel that appears active in the data, but in actuality part of the tissue is not active.

5. What are large vessel effects, and why do they matter for fMRI?

The BOLD resopnse deals with the deoxygenated hemoglobin in the blood.Within the vascular system that produces the BOLD signal, there are different components to consider. Deoxygenated molecules are paramagnetic and create gradients into the surrounding tissues. The primary mechanism for the BOLD signal is the dephasing of nucleur spins of water molecules as they diffuse through the gradients created by the paramagnetic deoxygenationed hemoglobin. Spins within blood vessels themselves give rise to a intravascular component, while spins located in the surrounding tissue produce an extravascular componet . In fMRI, the BOLD response includes both of these signal types. These signals can come from capillaries that are right next to the active neuron, or they can come from the deoxygenated hemoglobin that is removed from the from the brain by the venuous system, these signals are not really close to the active neuron, and therefore limit the spatial resolution of the BOLD response. Large-vessel effects refer to this phenomen that distant regions can give rise to a BOLD response.

6. What factors, aside from voxel size, influence spatial resolution in fMRI?

-Spatial scale: the exent of the region you are analyzing. If you are analyzing the brain overall, you will increase your spatial scale, while if you are interested in a sub-region of the brain, you can decrease you spatial scale to gather more relevant data to the task at hand.

-Smoothing: reduces the spatial accuracy, but makes the data more statistically significant

-Normalization: further reduces spatial accuracy, but provides a standardized image that helps in being able to make generalizations

-Use of ROI analysis: again spatial resolution diminishes because instead of looking at single voxels, groups of voxels are the focus, however this may increase functional resolution because of signal averaging becuase randomly distributed noises will be reduced.

7. What happens to estimates of the hemodynamic response as repetition time (TR) is reduced from very long (i.e., 4 s) to very short (i.e., 500 ms)?

Decreasing repetition time allows for more instances of measuring the changes of the hemodynamic response. As one looks at shorter intervals of what is happening temporally in the hemodynamic response, the spatial resolution increases and the data becomse more accurate.

8. Is there a preferred TR for fMRI? Does this depend on whether the design is event-related or blocked?

- There are basic preferred TR for fMRI studies (1-2s); event-related studies benefit from shorter TR but only up to a point; blocked studies benefit from longer TR

9. What are the disadvantages of high temporal resolution in fMRI?

-limiting spatial coverage, reducing signal amplitude, reduction of number of trials obtained for each condition. The amplitude of transverse magnetization following ecitation will be reduced at short TR's and less MR signal will be measured. Less slices can be acquired during a shorter time frame, therefore less spatial informaiton can be gathered.

**10. What is interleaved stimulus presentation? (Note: not interleaved slice acquisition.)**

A method to increase temporal resolutions, where experimental stimuli are presented at different points within a TR on different trials. It increases the effective sampling rate of an experiment at the expense of fewer trials per condition. If the events are presented at varied intervals, then the acquisition of data at those different intervals will give insight into different points in the hemodynamc response, thus more points will be gathered on the averaged hemodynamic response, therefore increasing temporal resolution.

11. Which is easier to study using fMRI, absolute event timing or relative event timing? Why?

It is easier to study relative event timing, because of the understanding of the temporal relation between the different regions of the brain. Though they won't get an immediate HDR they can understand the relation between the responses of the activation of the other sections of the brain.

12. How have small event timing differences been measured using fMRI? What are some caveats for such studies?

They measured with a gradient-echo ech-planar fMRI at high field (4 T) and high temporal resolution (TR of 100 ms), using hemifields. Hemifields were projecting to half a visual display and they were either shown at the same time or one before the other in order to see if the correspondence between stimulus introduction and brain activation was in correspondence between the two sides. From this the examined the different areas of the brain for their activation times. The authors investigated the differences between visual cortex and supplementary motor reaction time. They concluded that reaction times for these brain regions vary. The caveat to this is that differences in vascular properties contribute to the timing of activity, meaning that unless one isolates brain functions, overall response time may be the summed response of many different regions, which may skew your conclusion.

13. Name and define the two properties of a linear system.

Scaling: the principle that states that the magnitude of the system output must be proportional to the system input

Superposition: the principle of linear system that states that the total response to asset of inputs is equivalent to the summation of the independent responses to the inputs

14. How well is the fMRI response to a long-duration stimulus (e.g., 12 s) predicted by that to a short-duration stimulus (e.g., 3 s)?

The 12 s stimuli indicated linear superposition, but when reducing the stimuli to 3s, the response was larger than expected. This is tought to be caused by the neuronal adaptation effects, which is when the neurons decrease in activity over the first few seconds of a stimulus.

15. Why did Dale and Buckner describe the fMRI response as roughly linear?

When testing the hemodynamic responses evoked by increasing amounts of stimuli, they saw the the fMRI BOLD response adds across stimuli in a "roughly linear" manner in intervals similar to the intervals of experimental testing. So the principle of superposition held.

16. Approximately how long does the fMRI refractory period last?

Approximately 6 s.

Refractory period: a time period following the presentation of a stimulus during which subsequent stimuli evoke a reduced response.

17. How do refractory effects change the amplitude and latency of hemodynamic response?

The amplitude is reduced and the latency is delayed. For instance, at 1 s difference between the preceding stimulus, the amplitude was down 40% and the latency was 1 s late. For 2 s, the amplitude was down slightly more than 20% and around a second late as well.

18. How might one use the refractory effect to study neuronal adaptation? What information do these sorts of studies provide about brain function?

By understanding adaption, if we alter aspects of the stimuli presented we can understand what aspects of the stimuli are being adapted to. For instance, if we are given a face stimuli and certain aspects of the face are altered (size, position in sight, etc...), and there is no real change in response, we understand what aspects are being adapted to. This likewise can show that other aspects (such as lighting and view-points) which provide new neuronal stimulation, are not included in the process of adaptation in the area examined.

Chapter 9 focus on following:

Define ‘signal’ and ‘noise’ (q1)

Signal is the meaningful changes in some quantity. In fMRI, an important class of signals includes changes in intensity associated with the BOLD response across a series of T2* images. On the other hand, noise is the non-meaningful changes in quantity. In fMRI, there are many sources of noise and some can be classified as either noise or signal, depending on the study.

Describe different types of signal to noise (q2)

-Raw signal-to-noise-ratio is the ratio between MR signal intensity and the intensity of the thermal noise that is measured outside the sample. When designing an MR machine, engineers strive to minimize this ratio to create an optimally functioning MR unit.

-Contrast-to-noise-ration is the magnitude of the intensity difference between different quantities divided by the variability in their measurements. CNR is used to identify differences between tissues. Depending on the resolution of the CNR, the ratio between different tissue samples can be found. When CNR increases, the variability increases, but the difference between tissues is more greater.

-Functional signal-to-noise-ratio is the ratio between the intensity of a signal associated with changes in brain function and the variability in the data due to all sources of noise. CNR depends upon the intensity difference between voxels, whereas fSNR depends upon the intensity difference within a voxel, or group of voxels over time. fSNR is a main concern in fMRI studies because it shows the difference over time of a given region.

What are common forms of physiological noise (q7)

Physiological noise is commonly attributed to muscles contracting with each breath and heartbeat of the patient. Also, blood pulses through arteries and veins in the subject. The metabolic demands on neurons that drive chemical reactions needed to sustain life can create this physiological noise. Also, the subject shifting position or swallowing can create this. However, cardiac activity is the main form of physiological noise.

How does the balance between different types of noise change with field strength (thermal and physiological) (q15)

Thermal noise scales linearly with the field strength, so a 2.0-T scanner measures 2 times as much thermal noise as 1.5-T scanner. When one divides the quadratic increase in signal by the linear increase in noise, find that the raw SNR only increases linearly with the field strength. While thermal noise increases linearly with increasing field strength, physiological noise increases quadratically with the field strength. So, as field strength increases from 1.5-T to 3.0-T, raw signal will quadruple, thermal noise will double, and physiological noise will quadruple. Physiological noise may become dominant and the improvement of functional SNR with increasing field strength may be considerably less than linear. This implies that increased functional SNR has an increased spatial extent of activation.

Chapter 10 focus on the following

What is preprocessing vs. experimental analysis? (q1)

Preprocessing is the computational procedures that are applied to fMRI data following image reconstruction but before statistical analysis. Preprocessing steps are intended to reduce variability in the data that is not associated with the experimental task and to prepare the data for statistical testing. Experimental analysis is the analysis done after preprocessing steps are taken to reduce variability in the data such as head movement and other artifacts in the image.

What is the first rule of quality assurance? (q2)

The first rule is to examine your data. Many common artifacts are readily visible in the raw images even under cursory examination. You may do this as a time-series movie. This method of viewing the data may seem crude, but the human visual system is well equipped at picking up change over time, and with the time-series movie of MR data, one would be able to notice any glitches and big errors.

Why do we do slice timing corrections? (q4)

Each slice is acquired at a different time point within the TR (repetition time). Through small plane motions, as when a voxel moves from slice 12 to slice 13, the timing of activity will be off by one half of the TR value at affected time points. Thus slice timing correction first for interleaved sequences with long TR's will minimize timing errors.

How do researchers correct for head motion (q9)

For motion correction, successive image volumes in the time series are coregistered to a single reference volume. Because the brain is the same in every image of the time series, a rigid-body transformation is used. Rigid body transformation assumes that the size and shape of the two objects to be coregistered are identical and superimposes them onto each other through translations and rotations. Spatial interpolation, the estimation of intensity of an image at a special location that was not originally sampled using data from nearby locations, would also be done. This is done so that you can estimate the values that would have been obtained had there been no head movement.

Why do coregistration? (q11)

Coregestering allows us to map our functional data onto high-resolution and high-contrast structural images from the same subject. We would want to do this for 2 reasons. The first occurs when the two types of images were acquired at different locations, either because different slices were wanted for each or because the subject moved slightly between their acquisitions. A second reason for function-structural coregistration is image distortion.

Why do normalization? (q13)

Spatial normalization allows us to compensate for shape differences of the brain by mathematically stretching, squeezing, and warping each brain so that it is the same as every other brain. This allows combination of data across individuals.

What are the problems with normalization? (q16)

Normalization is based upon gross morphological features of brains and these features vary among individuals. Furthermore, these gross features do not necessarily indicate functional divisions between brain areas. Even when using cytoarchitecture instead of gross anatomy, boundaries between cytoarchitectonically distinct regions are highly variable.

Why smooth? (q 17)

BACK TO TOP

Because all subjects’ brain differ from one another in shape and size, and potentially functional organization, areas of activity are rarely represented in exactly the same voxels. Instead, combining data across many subjects distributes activity across a range of voxels and improves SNR. Smoothing not only increases the signal to noise ratio of each voxel, it also reduces the number of resolution elements that are assumed to be independent and used for correction of multiple testing. Spatial smoothing and signal averaging both serve to increase the SNR. Spatial smoothing, however, combines data from multiple voxels, instead of data from multiple trials, in order to get a more robust SNR.

Explain the problem of multiple comparisons in brain imaging (q18)

The multiple comparison problem is the increase in the number of false-positive results with increasing number of statistical tests. It is of particular consequence for voxelwise fMRI analyses, which may have many thousands of statistical tests. In short, if you set the significance at alph < .05, then there should be more than 5000 voxels active due to chance alone because a typical functional imaging volume has more than 100,000 voxels. If you don’t correct for multiple comparison, you will get random significance.

What is temporal filtering? Why do low pass filtering? Why do high pass filtering? (p21)

Temporal filtering allows one to selectively attenuate frequencies around a certain range while keeping intact other frequencies. This is useful when one wants to eliminate certain variabilies such as heart rates. Typical heart rates during an fMRI experiment vary, but are often between 1 to 1.5 Hz. For comparison, a typical experimental design might present alternating blocks of 12s of task and 12 s of rest, for a total presentation rate of .04 Hz. One can use a low pass filter to exclude frequencies above .2 Hz to remove physiological oscillations such as heart rate.

Chapter 11: Experimental Design

1. Define an “experiment” in your own words.

An experiment is the designing and implementation of a manipulation of a variable which is performed in order to evaluate a hypothesis. Usually experiments are aimed at adding to generalizable knowledge.

2. What are the differences between independent and dependent variables?

An independent variable is the aspect of the experiment that researchers manipulate. By changing the independent variable, the researcher presumes that the dependent variable (what is being measured) will also change.

The dependent variable in a study is the quantity that is measured in order to determine the effect of changing the independent variable.

3. What is a research hypothesis? What three basic types of hypotheses are possible for fMRI studies, and what are their characteristics?
BACK TO TOP

A hypothesis is a statement about the nature of the world that makes predictions about the results of an experiment. A scientific hypothesis must be able to be proven false, or else it is not amenable to scientific investigation.

In fMRI there are hypotheses about:

1) hemodynamic activity, which examine the BOLD effect itself, without addressing the question of causation;

2) neuronal activity, which is estimated by transforming the measured BOLD signal, as neuronal activity cannot be measured by fMRI directly. These types of hypotheses use the BOLD signal to infer neuronal activity.

3) psychological hypotheses, which rely upon broader concepts that can be difficult to define, and are therefore difficult to formulate, but are also influential in forming theories.

4. Why do some people consider fMRI data to be epiphenomenal? What do you think about this issue?

An epiphenomenon means that something (in this case, BOLD activity and fMRI data) is a result of some side-effect or primary phenomenon, instead of resulting from the presumed phenomenon (in this case, neuronal activity). In other words, correlation does not entail causation, and some people think that BOLD activity and neuronal activity are not causally related. I think this is a criticism that deserves more attention, because there are areas to explore that could at least help to settle some of this debate.

5. What are confounding factors in an experiment, and how can they be minimized?

A confounding factor is a factor that changes along with the independent variable and makes attribution of changes in the dependent variable more complicated (because attributing change in the dependent variable to the change in the independent variable does not eliminate the possibility that the confounding factor actually caused the change). To minimize confounding factors, experimenters should ensure that randomization is complete (i.e. ensuring that factors that vary within the experiment change in a random way with respect to the independent variable). Experimenters can also use a technique called counterbalancing, to equalize the influence of confounding factors across conditions of the independent variable.

6. What are the basic principles of blocked designs? Why are they sometimes referred to as “subtractive” designs?

A blocked design compares an experimental condition that is assumed to be present at all times for a defined period of time to a control condition or a different experimental condition. Basically, an experiment is divided into blocks of time, and the conditions within each block alternate. The data can be easily analyzed by comparing the dependent measure in each block condition. They are referred to as subtractive because conditions are compared to each other by analyzing which voxels are active in one condition that are not active in another condition. In other words, BOLD activity differences between the two conditions are determined by subtracting one condition from the other.

7. Are blocked designs better for detection of activity or estimation of the time course of activity? Why?

In general, they are better at detection of activity, because within a single block (a single experimental condition) the processes may actually be very different and change over time. However, a blocked design only compares the blocks to other blocks. Also, the shape of the hemodynamic response is not examined as closely in blocked designs as in event-related designs, which means that the order in which voxels showed activity cannot be determined from the data.

8. Why do some experiments evoke increased activity during control conditions compared to experimental conditions?

What might subjects be doing/thinking during control blocks? It is possible that in certain experiments, the experimental condition involves inhibition rather than activation of certain processes, so that less activity in some areas of the brain occurs during the experimental task. Some researchers have speculated that introspection during ‘rest’ conditions account for the increase in activity of some areas.

9. What physiological measure did Gusnard and Raichle suggest as an index of baseline activity in the human brain?

The oxygen extraction fraction, which is the proportion of oxygen extracted from the blood. This measure remains stable across brain regions, so that the proportion of oxygen extracted during a baseline condition is basically spatially uniform.

10. Which brain regions show increased activity during control conditions?

The lateral parietal cortex (angular gyrus), the posterior cingulated or precuneus, the superior frontal gyrus, and the ventral prefrontal cortex

11. What are the basic principles of event-related designs?

Event-related designs assume that neural responses to stimuli in the experiment will be brief and discrete.

Event-related designs measure responses in the brain over time, as opposed to steady state brain activity measured in blocked designs. The detection power and temporal resolution are the key factors in event-related design, while spatial resolution is less important. The hypothesized duration of the hemodynamic response is more important in this type of design than in others.

Each stimulus is considered an impulse, which generates a hemodynamic response. These responses are assumed to conform to the linear systems model, meaning that BOLD responses to successive stimuli do not interact with one another.

12. What does the term “epoch” describe in an event-related design?

An epoch is a segment of an overall time series or set of images that is separated from the full series and time-locked to a particular event. It usually is a period surrounding the event of interest (before, during, and after).

13. How do researchers often improve signal-to-noise in event-related designs?

By adjusting the interstimulus intervals.

14. What is the advantage of a prestimulus baseline period for event-related analyses?

This allows analysis of the differences between two experimental conditions, as well as comparisons of each condition with the baseline. Otherwise only the relative difference between the two experimental conditions could be analyzed, which may not provide much information about the process overall.

15. What is the primary disadvantage of slow event-related designs?

Their inefficiency, due to the low density of events over time (requiring longer scan).

16. What sorts of experimental questions can be answered by event-related designs, but not by blocked designs?

In general, event-related designs have better estimation power, and is better able to characterize the timing of events in the brain (as opposed to spatial properties). Event-related designs are important for questions about the timing of the involvement of brain areas for a task, and are helpful in identifying cognitive processes associated with distinct time periods. They also allow trial sorting, unlike blocked designs.

17. What is trial sorting? In what sorts of experiments would it be useful?

Trial sorting refers to the analysis of data in a different way than was originally planned for—in other words, after the experiment has been carried out, one can examine the same data by reassigning events to condition. This is useful in experiments that provide additional sources of data, for instance when behavioral data are collected along with imaging data. It is also used to analyze data based on the responses of subjects, such as error rates or response time.

18. What are semirandom designs? What advantages do they provide

A semirandom design is one that varies the probability of the events or stimuli that occur. It is a type of event-related design that uses fixed intervals, but within those intervals the events occur randomly, so that in one interval many stimuli may occur while in the next only one may occur. This design can optimize the detection and estimation power, which are at odds with each other, and should be used in studies that value both detection and estimation equally.

19. What are mixed designs? Why would they be used?

A mixed design combines blocked and event-related designs into a single study, typically by having blocks within which multiple types of events occur instead of only one condition per block. A mixed design allows statistical analysis of independent variables that do not co-vary. Each block is associated with a broader shift in strategy (e.g. attention), and analysis that compares blocks can thus measure a state-related process. A mixed design is therefore chosen when a sustained change in brain activity is the target of the investigation.

Event-related designs, and the short-term changes they induce in the brain, contain individual events that can measure item-related processes, which are changes associated with a particular stimuli. When research questions involve both types of processes, studies can be designed to incorporate both state and item-related processes.

Chapter 12: Statistical Analysis

1. What is the difference between descriptive and inferential statistics? Which do we use to assess statistical significance in fMRI?

Descriptive statistics are limited to describing features of the particular data set in question (e.g. the mean, median, etc.). Inferential statistics are used to determine what the data collected from a sample of a population can indicate about that population as a whole (i.e. what we can infer from the data about the limited group to the larger group in general).

2. What are Type I and Type II errors? Which is typically minimized in fMRI data analysis?

Type I error is rejecting your hypothesis when it is really true. In the case of fMRI it would be a voxel labeled as active when it really is not. Type II error is the opposite, retaining your hypothesis when it is false. Type I errors are minimized in fMRI.

3. What is an alpha value? Does it relate to Type I or Type II errors?

The alpha value is the arbitrary threshold set to determine if data is statistically significant. If the obtained probability for the sample is less than the alpha value the sample is said to be statistically significant and the hypothesis is rejected. Yes the alpha value relates to type I and II errors. As the alpha value decreases you decrease the probability of making a Type I error, but increase the probability of making a Type II error.

4. Describe the basic principles of a t-test. In what sorts of analyses are t-tests most commonly used?

The t-test compares two distributions and identifies differences in the means of the samples. It is most commonly used in blocked designs to compare two conditions, although it may be used for certain event-related designs.

5. What are the advantages and disadvantages of the Kolmogorov–Smirnov test compared to the t-test? Why is it infrequently used?

The K-S test is a tests for statistical significance that evaluates whether two samples are drawn from the same distirbution.The K-S test has the advantage of being sensitive to changes in variability and shape in sample means. The disadvantage is that the K-S test is less sensitivity to changes in the sample means. The K-S test is not frequently used because fMRI studies are usually not looking for variability caused to the dependent variable, but a change in the sample means. i.e. brain has more activity when doing this- not more variability. A t-test is pirmarily used for identifying the differences in the means of two samples of data. This test is applicable if a research question focuses on specific data points, a specifc region of time (epoch), within an event related design. The t-test is good for comparing means, but not for looking at the variability or shape. The K-S test converts each distribution to a cumulative distrubution function, which plots the proportion of the data at or below each possible value of the dependent variable. T-test should be used when manipulation is going to have an additive effect on the dependent variable, while K-S test should be used when manipulation is expected to have an effect on the variability of the dependent variable.

6. What are the basic principles of a correlation analysis? Over what range do correlation coefficients vary?

Correlation analysis is a statistical test, which evaluates the strength of the relationship between two variables on a scale from -1 to 1. A correlation of -1 means the objects are negatively correlated, as one increases the other decreases. The closer to 1 or -1 the stronger the relationship between the variables is. Correlation analysis allows comparision between the observed and predicted hemodynamic response.

7. What are the effects of signal averaging upon a correlation analysis?

Signal averaging can reduce the correlation found in correlation analysis. Low-frequency changes in the data caused by scanner drift can cause the correlation between the data and predicted to decrease in strength.

8. What does a Fourier transform do to a time series of data?

It converts a time series into a frequency series, which means that it only works on experiments (e.g. alternating blocked design), that have some sort of regularity or frequency in its pattern of conditions. Data transformed into a power spectrum displays the frequencies of the data.

9. What is the Nyquist Sampling Theorem? Why is it important for fMRI?

The Nyquist sampling theory states that to accurately measure a given frequency, you must sample at a minimum of twice that frequency.

10. What are some different ways in which fMRI data can be displayed? What are their advantages and disadvantages?

The most common way fMRI data is displayed is in a single anatomical slice, which shows activity through the use of color (generally, bright = high significance, dull=low). The advantages for a single slice display are activity is easily seen and they require little processing after the experiment. The disadvantages with single slice are it is difficult to identify activity in gyri and sucli because of the variability in the individual brain. Additionally, determining which single slices to use for experimental data is difficult. Another type of fMRI display is a rendered image, which converts the 2D images from the experimental data into a 3D image. In rendered images activity is easily identified, even in the sulci and gyri. A disadvantage to rendered images is they do not show internal nuclei, so activation of internal structures is not seen. Researchers get around this by using glass brain views, which present fMRI data in 2D, as if the brain is transparent and only activation is shown. Glass brain views are present in three images to zone in on the area of activation

11. What is the difference between radiological and neurological conventions for displaying MRI data?

In radiological convention, the left side of the image corresponds to the right hemisphere of the subject. In neurological convention, the left side of the image corresponds to the left hemisphere of the brain.

12. What are the principles of the General Linear Model (GLM)? How do we evaluate the significance of activity using the GLM?

The GLM models observed data as a function of a linear combination of multiple independent factors plus a constant (noise). Linear refers to the notion that they can simply be summed, that they do not influence each other (e.g. the effect of a particular variable would be the same with or without the others). The GLM assumes that the hemodynamic response is linear.

13. What is a design matrix?

A design matrix is used in GLM and specifies how the linear model changes over time. It is a hypothesis about which factors are causing the variance in the measured data.

14. What are nuisance factors, and why might they be included in an analysis model?

Nuisance factors are additional factors within the design matrix that are associated with known outside sources of variability. Nuisance factors are included because they can reduce the amount of residual variability. Additionally, by using nuisance factors, experimenters increase the validity of their GLM because sources of variability are being identified and accounted for.

15. What assumptions does the GLM make? How valid are these assumptions for fMRI?

One assumption of GLM is the use of the same design matrix throughout the brain, which can decrease region variability, but increase residual error. This assumption as said can cause problems with data analysis, but can be overcome by combining GLM with a region-of-interest analysis. A second assumption is all voxels are analyzed independently. Thirdly, is all time points are independently distributed, so the residuals will be similarly distributed. This assumption can become a problem when the subject moves or when there is scanner drift. Finally, design matrixes contain factors that accurately reflect changes in BOLD due to neural activity. This assumption is somewhat valid, but a problem arises when the neural response takes time after the stimulus to occur because in most data the stimulus and any neural changes directly after the stimulus is applied are grouped (when really the neural change from the stimulus happens seconds later).

16. What are data-driven analyses? What advantages and disadvantages do they present for fMRI?

In data-driven analyses, experimenters search the structure of the data in hopes of finding task-related activations. An advantage of data-driven is it is not as dependent on hemodynamic response as traditional analysis, but grouping the data into clusters creates a new problem: How many clusters do you have in a data set? Another advantage of data-driven is it has been shown to yield similar results as hypothesis driven analysis. Finally, a disadvantage of data-driven is it can increase the intersubject variability. A type of data-driven analysis is independent components analysis. This analysis identifies spatially stationary sets of voxels whose acticity caries together over time and is maximally distinguishable fro that of other sets. Data-driven analysis would be used when the underlying hemodynamic response is not known and the researcher is trying to find patterns in activation that will suggest this information.

17. Why is random field theory used to estimate the number of independent tests in fMRI analyses? What are its principles?

RFT is used to estimate the number of independent tests because it is based on the smoothness of the experimental data, which is the degree to which neighboring voxels are temporally correlated.

18. What are the advantages and disadvantages of cluster size thresholding?

An advantage of cluster size thresholding is it reduces the amount of false positives through increasing the cluster size. Also, by reducing the alpha value, cluster-size threshold decreases the number of Type II errors or missing a true activation. One disadvantage of it is cluster-size looks over a large area and thereby looking over the areas of smaller activation, which could be just as meaningful. Another disadvantage it will over look nonspeherical areas. A final disadvantage is cluster-size does not correlate activity among adjacent voxels. It assumes all voxels have the same likelihood of being active, which is false.

19. What are region-of-interest (ROI) analyses? Why might a study use ROI analyses instead of voxelwise analyses, or vice versa?

ROI analysis is a process for evaluating statistical data in regions that are predetermined, usually before the experiment, and usually based on anatomical differences. ROI could be used over voxelwise because it reduces the number of comparisons, but it is difficult using ROI because it is difficult to find a way to create the separate regions. Additionally, ROI present the advantages of reducing the problems in comparing subjects because ROI are individually created and do not becomie skewed by normalizing. Once a region of interest is created, the same region can be found in multiple people regardless of whether it is in a slightly different location.

20. What is the difference between fixed-effects and random-effects analyses? Which is considered more appropriate for generalizing fMRI results to the population from which the subjects were drawn?

Fixed-effects allow inferences to be made about the particular subject in the experiment, while random-effects allows for inferences to be made about the population. Random effects analysis is considered more appropriate for fMRI research because it deals with making inferences on the population.

21. What challenges must be overcome in order to use fMRI as a diagnostic tool for presurgical patients?

In order for fMRI to be used as a presurgical tool, the creation of data and the analysis of it must be done quickly and accurately. Also, the interpretation of the significant activations needs to be refined and become more accurate.

Additional study

You should be able to draw the bold response due to the following. Graphs should have correctly labelled x and y axes with approximately correct times and % BOLD modulation values. (covered in Ch7, q 18,19,20; Ch8)

(excel version of graphs would be excellent!)

1) A brief event (0.5 sec duration)

2) Two brief events separated in time by 2 secs (draw both summed and separate responses)

3) Two brief events separated in time by 6 secs (draw both summed and separate responses)

4) A repetitive sequence of events (>10) occurring over a time period of 15 secs. E.g. a block of events

Comments (0)

You don't have permission to comment on this page.

Anonymous Create an account or Log in

To join this workspace, request access.

Already have an account? Log in!

Loading…

Current Homework

Certain groups will read certain chapters for discussion during the next class.

[Update this SideBar every week, so everyone knows what they have to do for next class!]

Loading…

fmri

Complete Study Guide

Chapter 1

1. Describe the goals and methods of phrenology. What concept did the phrenologists introduce?

2. What is functional magnetic resonance imaging? BACK TO TOP

3. What are the limitations of lesion studies of the brain? How can functional neuroimaging help overcome these limitations?

4. What are the limitations of drug studies of the brain? How can functional neuroimaging help overcome these limitations?

5. What are the limitations of electrophysiological studies of the brain? How can functional neuroimaging help overcome these limitations?

6. To what aspect of imaging does “contrast” refer? How could a single image be high-contrast in one sense and low-contrast in another? BACK TO TOP

7. What is the difference between structural contrast and functional contrast?

8. What are voxels?

9. What is functional resolution, and how is it different from spatial and temporal resolution?

10. What is resonance?

11. Why did physicists use oscillating magnetic fields to study magnetic resonance effects?

12. Describe the experimental apparatus used by Felix Bloch and his colleagues to measure nuclear magnetic resonance effects.

13. How did Damadian and Lauterbur each contribute to the development of MRI?

14. What is an image? What advance was most critical to the development of techniques for image formation in MRI?

15. What were the two reasons for the change in terminology from “nuclear magnetic resonance” to “magnetic resonance imaging”? BACK TO TOP

16. Why was there a boom in MRI use in the 1980s? How did this growth set the stage for fMRI?

Chapter 2

1. What are the three main components of an MRI scanner?

2. How does an MRI scanner generate the main magnetic field? What two criteria are important for the main magnetic field?

3. Why are superconducting electromagnets necessary for MRI?

4. What is the difference between a surface coil and a volume coil?

5. Why are gradients necessary for image generation? What sorts of coils are used to generate these gradients?

6. What is shimming, and why is it important?

7. Why might researchers want to monitor physiological changes like cardiac and respiratory rate during an fMRI experiment?

8. Describe the procedures of a typical fMRI experiment, beginning with recruitment of the subject.

9. What sorts of conditions/problems would prevent someone from being a subject in an fMRI experiment?

10. What effects do very strong static magnetic fields have upon human tissue?

11. Forty five percent of subjects reported unusual sensations when entering the bore of a 4-T scanner. What was most important about this result?

12. What happens to metal brought within the static magnetic field? Consider both large external objects (e.g., oxygen canisters) and small internal devices (e.g., aneurysm clips).

13. What effects do the changing gradient fields have upon the human body? How can these effects be minimized?

14. What is SAR? Why is it important for fMRI?

15. Why is it important to avoid looping wires or necklaces near the head coil?

16. What is the most common health consequence for MRI studies? How can it be minimized?

Chapters 3, 4, 5

What is resonance (simple explanation)? What determines resonance in MR and how is it used to select slice. (Ch3, q5, 8, 10, 12)

BACK TO TOP

How is location determined within slice (e.g. see ch4 q 7, 16)

Be able to explain difference between T1 vs. T2 (ch3, q14-16; ch 5, q6-10)

What is a fourier transform?

How does k-space differ from ‘normal space’ illustrate this with some concrete examples (see ch4, q’s 5, 11, 12, 14)

Why are slices interleaved, how does this affect temporal characteristics of fMRI? (for first part, see Ch4, q15)

What is TR and TE (Ch5, q3,4)

What is a susceptibility artifact? (Ch5, q18)

Chapters 6 & 7

Focus on location in the brain, and ability to actually recognize and label views of the brain (q’s 9-12, but make sure you know what these look like!)

What influences blood flow? (Ch6, q18)

What magnetic properties is the BOLD sensitive to? (Ch7, q1)

What does BOLD stand for? (Ch7, q 6)

What causes BOLD contrast? (Ch7, q7)

What influences spatial resolution of BOLD? (Ch7, q9)

Why might BOLD be thought of as a quirk? (Ch7, q 12)

What special type of BOLD effect might have better spatial resolution, but has yet to be clearly seen (Ch7, q13)

Chapter 8: Spatial and Temporal Properties of fMRI

1) What are the major divisions of fMRI data, from subjects down to voxels?

2. What are disdaqs, and why are they sometimes incorprated into imaging protocols?

3. What are the disadvantages of high spatial resolution in fMRI?

4. What are partial volume effects?

5. What are large vessel effects, and why do they matter for fMRI?

6. What factors, aside from voxel size, influence spatial resolution in fMRI?

7. What happens to estimates of the hemodynamic response as repetition time (TR) is reduced from very long (i.e., 4 s) to very short (i.e., 500 ms)?

8. Is there a preferred TR for fMRI? Does this depend on whether the design is event-related or blocked?

9. What are the disadvantages of high temporal resolution in fMRI?

10. What is interleaved stimulus presentation? (Note: not interleaved slice acquisition.)

11. Which is easier to study using fMRI, absolute event timing or relative event timing? Why?

12. How have small event timing differences been measured using fMRI? What are some caveats for such studies?

13. Name and define the two properties of a linear system.

14. How well is the fMRI response to a long-duration stimulus (e.g., 12 s) predicted by that to a short-duration stimulus (e.g., 3 s)?

15. Why did Dale and Buckner describe the fMRI response as roughly linear?

16. Approximately how long does the fMRI refractory period last?

17. How do refractory effects change the amplitude and latency of hemodynamic response?

18. How might one use the refractory effect to study neuronal adaptation? What information do these sorts of studies provide about brain function?

Chapter 9 focus on following:

Define ‘signal’ and ‘noise’ (q1)

Describe different types of signal to noise (q2)

What are common forms of physiological noise (q7)

How does the balance between different types of noise change with field strength (thermal and physiological) (q15)

Chapter 10 focus on the following

2. What is functional magnetic resonance imaging?
BACK TO TOP

6. To what aspect of imaging does “contrast” refer? How could a single image be high-contrast in one sense and low-contrast in another?
BACK TO TOP

15. What were the two reasons for the change in terminology from “nuclear magnetic resonance” to “magnetic resonance imaging”?
BACK TO TOP

**10. What is interleaved stimulus presentation? (Note: not interleaved slice acquisition.)**

3. What is a research hypothesis? What three basic types of hypotheses are possible for fMRI studies, and what are their characteristics?
BACK TO TOP