QUIZ QUESTIONS ON PULSE SEQUENCES

Pulse Sequences Quiz

MR SIGNALS

  1. The MR signal generated by a single RF-pulse is called a
    1. Free induction decay (FID)
    2. Spin echo (SE)
    3. Gradient echo (GRE)
    4. Stimulated echo (STE)

    The FID is produced by a single RF-pulse as the tipped nuclear magnetization precesses around Bo and generates current in the receiver coil via induction (Faraday-Lenz Law). Link to Q&A discussion

  2. Concerning the FID, which of the following is true?
    1. It oscillates indefinitely at the resonance frequency
    2. It decays with time constant T2*
    3. It increases in amplitude with time constant T1
    4. It is generated by a perfect 180º pulse

    The FID is a damped sinusoidal oscillation at the resonance frequency that decays with time constant T2*. Link to Q&A discussion

  3. If a magnetic field gradient is turned on immediately after an RF pulse, what happens to the MR signal?
    1. It becomes a free induction decay (FID)
    2. It remains unchanged
    3. It dies out quickly
    4. It becomes refocused into a gradient echo (GRE)

    The initial MR signal is already synonymous with the FID (choice a is false). Turning on a gradient during the FID causes it to decay more quickly than it would by T2* processes alone (choice c is correct). A second, later applied gradient with opposite polarity is required to generate a GRE. Link to Q&A discussion

  4. Another name for a gradient echo is a
    1. Field echo
    2. RF echo
    3. Spin echo
    4. Stimulated echo

    Less commonly used, field echo (FE) is the correct answer. Some manufacturers, including Philips and Canon, use FE or FFE (fast field echo) to name their gradient echo sequences. Another synonym is gradient-recalled echo (GRE). Link to Q&A discussion

  5. What happens if a gradient is turned on after the FID has spontaneously decayed?
    1. A gradient recalled echo is produced.
    2. A spin echo is produced.
    3. The FID is partially regenerated.
    4. Nothing.

    Once the FID has spontaneously decayed by T2* processes, no signal can be recovered by action of a gradient. Link to Q&A discussion

  6. The first lobe of a gradient applied during a simple GRE imaging sequence is called the
    1. Readout gradient.
    2. Dephasing gradient.
    3. Rephasing gradient.
    4. Bipolar gradient.

    The first step in a GRE sequence is to dephase the FID, so this lobe is referred to as the “dephasing gradient”. The dephasing gradient results in accelerated dephasing beyond normal T2* processes with 'squelching/ scrambling' of the FID. Link to Q&A discussion

  7. The rephasing gradient refocuses spins that have been dephased by
    1. The dephasing gradient only
    2. T2 only
    3. T2* only
    4. The dephasing gradient, T2, and T2*

    The rephasing gradient only refocuses spins scrambled by the dephasing gradient itself. T2 and T2* processes are unaffected and contribute to continued signal loss Link to Q&A discussion

  8. Spin echoes were first described by
    1. Felix Bloch
    2. Edward Purcell
    3. Isador Rabi
    4. Erwin Hahn

    Erwin Hahn discovered spin echoes while a graduate student in 1949. Link to Q&A discussion

  9. A single spin echo is produced by
    1. A single RF-pulse
    2. A single RF-pulse plus gradient reversal
    3. Two RF-pulses
    4. Three RF-pulses

    A single RF pulse generates a free induction decay (FID), but two successive RF pulses produce a spin echo (SE). Link to Q&A discussion

  10. In a spin echo pulse sequence, the time between the middle of the first RF-pulse and the peak of the spin echo is
    1. TR
    2. TE
    3. TE/2
    4. TI

    The time between the middle of the first RF pulse and the peak of the spin echo is called the echo time (TE). Link to Q&A discussion

  11. In a spin echo pulse sequence, the time between the first and second RF-pulses is
    1. TR
    2. TE
    3. TE/2
    4. TI

    The spin echo occurs at time TE, which is twice the interpulse spacing. So the correct answer is TE/2 Link to Q&A discussion

  12. Spin-echoes for imaging are often produced using a 90º-pulse followed by a 180º-pulse. What would happen if the second pulse were changed to 90º?
    1. No echo would be produced.
    2. A normal sized echo would be produced at a time 2 x TE
    3. An echo half as large would be produced at time TE
    4. An echo half as large would be produced at time 2 x TE

    A 90º-90º RF pair was what Erwin Hahn originally used, and the resultant echo is often known as a Hahn echo. A Hahn echo will be formed from any two arbitrary RF pulses at time TE, with magnitude reduced depending on the angles chosen. The maximum SE intensity of a 90°-90° pair is only half as large as that produced by a 90°-180° pair. Link to Q&A discussion

  13. Which spin-echo pulse pair would be expected to produce the largest signal?
    1. 90º-90º
    2. 90º-180º
    3. 180º-90º
    4. 180º-180º

    The reason 90°-180° pairs are most commonly used is that this combination produces the maximum possible echo signal. The maximum SE intensity of a 90°-90° pair is only half as large as that produced by a 90°-180° pair. In general, if the first RF-pulse has flip angle α1 and the second has flip angle α2, the maximum signal intensity of the Hahn echo will be smaller than the conventional (90°-180°) echo by a factor of (sin α1)•(sin² α2/2). Link to Q&A discussion

  14. How many echoes may a series of 3 closely spaced successive RF-pulses produce?
    1. 3
    2. 4
    3. 5
    4. 6

    Three RF-pulses can produce up to 5 spin echoes: 3 primary, one secondary, and one stimulated. Link to Q&A discussion

  15. Of these, how many are considered to be “stimulated” echoes?
    1. 1
    2. 2
    3. None of them
    4. All of them

    Only 1 stimulated echo is produced by 3 RF pulses. Link to Q&A discussion

  16. A sample placed in a static magnetic field develops an initial longitudinal magnetization Mo. It is then subjected to series of RF-pulses equally spaced apart by time TR. If TR>>T1, what can be said about Mz, the longitudinal magnetization immediately before each subsequent RF-pulse?
    1. It cannot be predicted.
    2. It is the same as Mo.
    3. It is larger than Mo.
    4. It is smaller than Mo.

    If TR is sufficiently long (the usual requirement is TR > 4-5x T1), the longitudinal magnetization will have time to completely recover and be restored to its full initial magnetization Mo at the time of the next RF-pulse. Link to Q&A discussion

  17. A sample placed in a static magnetic field develops an initial longitudinal magnetization Mo. It is then subjected to series of RF-pulses equally spaced apart by time TR. If T2* < TR < T1, what can be said about Mz, the longitudinal magnetization immediately before each subsequent RF-pulse?
    1. It cannot be predicted.
    2. It is the same as Mo.
    3. It is larger than Mo.
    4. It is smaller than Mo.

    In this scenario, TR is now the same or shorter than T1. Here the second and subsequent pulses occur before the longitudinal magnetization has returned to its initial value (Mo). After a few pulses a new steady-state longitudinal magnetization (Mss) will be established, where Mss) < Mo. Link to Q&A discussion

  18. The above situation (in Question 17) is known as
    1. Partial saturation
    2. Complete saturation
    3. Steady-state free precession
    4. Incomplete magnetic equilibrium

    This phenomenon is called partial saturation, meaning that the spin system has not fully relaxed back to its equilibrium condition. Link to Q&A discussion

  19. A sample placed in a static magnetic field develops an initial longitudinal magnetization Mo. It is then subjected to series of RF-pulses equally spaced apart by time TR. If T2* < TR < T1, what can be said about Mxy, the transverse magnetization, midway between subsequent RF-pulses?
    1. It cannot be predicted.
    2. It is the same as Mo.
    3. It is smaller than Mo but not zero.
    4. It is zero.

    The sample is partially saturated, as noted in question #17, reaching a new non-zero steady-state longitudinal magnetization, Mss) < Mo. But because TR > T2*, complete decay of the transverse magnetization occurs midway between each pair of RF-pulses, so the answer is zero (d). Link to Q&A discussion

  20. Same scenario as Question #19, but here TRMxy midway between each pair of pulses?
    1. It cannot be predicted.
    2. It is the same as Mo.
    3. It is smaller than Mo but not zero.
    4. It is zero.

    In this situation with TR so short, the transverse magnetization Mxy does not fully decay between RF pulses, reaching a non-zero value less than Mo. Link to Q&A discussion

  21. The above situation is known as
    1. Partial saturation
    2. Complete saturation
    3. Steady-state free precession
    4. Complete magnetic equilibrium

    If TR is made even shorter (≤T2*) the FID following each RF-pulse does not completely die out before the SE/STE component begins to form. The tails of the FIDs and SE/STEs merge together. In this situation the transverse signal never completely disappears and the result is called a steady state free precession (SSFP). Link to Q&A discussion

  22. If the initial longitudinal magnetization of a sample is Mo, what is the longitudinal magnetization immediately after a 90º-pulse?
    1. Mo
    2. −Mo
    3. ½ Mo
    4. Zero

    After an ideal 90º-pulse the longitudinal magnetization should be zero. Link to Q&A discussion

  23. If the initial longitudinal magnetization of a sample is Mo, what is the transverse magnetization immediately after a 90º-pulse?
    1. Mo
    2. −Mo
    3. ½ Mo
    4. Zero

    After an ideal 90º-pulse all the longitudinal magnetization should be converted to transverse. Link to Q&A discussion

  24. If the initial longitudinal magnetization of a sample is Mo, what are the longitudinal (Mz) and transverse (Mz) magnetizations immediately after a 45º-pulse?
    1. About 90% of Mo
    2. About 70% of Mo
    3. About 50% of Mo
    4. About 10% of Mo

    From trigonometric identities we see that when a magnetization vector initially aligned with the z-axis is flipped by angle α, the transverse and longitudinal components of magnetization after the flip will be [Mo sin α] and [Mo cos α], respectively. For α=45º, sin 45º = cos 45º = 1/√2 ≈ 0.707. So about 70% of Mo is the correct answer (b). Link to Q&A discussion

  25. The flip angle that maximizes MR signal from a tissue in a GRE sequence without transverse coherences is known as the
    1. Hahn angle
    2. Bloch angle
    3. Purcell angle
    4. Ernst angle

    Named after Richard R. Ernst, who received the Nobel Prize in Chemistry in 1991 for his discovery. Link to Q&A discussion

  26. When is the Ernst (optimal flip) angle approximately equal to 90º?
    1. When TR >> T1
    2. When TR ≈ T1
    3. When TR << T1
    4. Never

    The Ernst angle (αE) can be calculated from the equation, αE = arccos (e−TR/T1). It reaches its maximum of 90º only when TR >> T1. Link to Q&A discussion

Back to top of page

SPIN ECHO AND INVERSION RECOVERY

  1. In a spin-echo pulse sequences the 180°-pulse refocuses spins dephased by
    1. Magnetic field inhomogeneities
    2. Diffusion
    3. T2 processes
    4. All of the above

    The 180°-pulse allows refocusing of nonmoving spins whose phases have been scattered by constant field distortions and inhomogeneities. The pulse does not correct for T1 or true T2 effects due to random processes at the atomic/molecular level. It does not correct for phase shifts of spins that move, flow, diffuse or undergo chemical exchange. Link to Q&A discussion

  2. Comparing Multi-echo Spin Echo (MSE) and Fast (Turbo) Spin Echo (FSE), which of the following statements is false?
    1. Both MSE and FSE use more than one 180°-pulses during each TR interval.
    2. MSE uses only one phase-encoding gradient during each TR interval.
    3. FSE applies the phase-encoding gradient multiple times during each TR interval.
    4. The echo amplitude decreases with successive 180°-pulses in both MSE and FSE imaging.

    In FSE different phase-encoding gradients are being applied with each 180°-pulse, so the amplitude of the echo may vary depending on the phase-encode step. In MSE the phase-encoding gradient is turned on only once in each TR interval, so successive echoes become reduced in signal intensity. Link to Q&A discussion

  3. Which of the following spin-echo parameter selections would produce a proton-density-weighted image?
    1. TR = 500, TE = 10
    2. TR = 500, TE = 100
    3. TR = 5000, TE = 10
    4. TR = 5000, TE = 100

    Answer c), a long TR and short TE are used to produce a proton-density-weighted image. Link to Q&A discussion

  4. A spin-echo sequence using a short TR and long TE produces
    1. A T1-weighted image
    2. A T2-weigthed image
    3. A PD-weighed image
    4. A noisy low contrast image

    The correct answer is d). So this combination is seldom used. Link to Q&A discussion

  5. Tissues or lesions with long values of T1 and T2 are typically
    1. Bright on T1-weighted SE images
    2. Bright on T2-weighted SE images
    3. Bright on both T1- and T2-weighted SE images
    4. Dark on both T1- and T2-weighted SE images

    Such lesions are typically bright on T2-weighted images and dark on T1-weighted images. Link to Q&A discussion

  6. By alternating the phase of 180°-pulse transmission, the CPMG (Carr-Purcell-Meiboom-Gill) modification of the spin-echo pulse sequence
    1. Reduces cumulative phase errors from repetitive imperfect 180°-pulses.
    2. Reduces artifacts due to B1 inhomogentity effects.
    3. Improves estimation of T2 values
    4. None of the above
    5. All of the above

    Developed in 1958, the CPMG technique is still used in multi-echo spin echo sequences for all the reasons listed. Link to Q&A discussion

  7. In the Inversion Recovery (IR) spin-echo pulse sequence, the inversion time (TI) is defined as
    1. The time between the first 180°-pulse and first 90°-pulse
    2. The time between the first 180°-pulse and the echo
    3. The time between the two 180°-pulses in each TR interval
    4. The time between the first 90°-pulse and second 180°-pulse

    Answer a) is the definition of TI. Link to Q&A discussion

  8. Advantages of IR techniques include
    1. Shorter scan times
    2. Decrease in flow-related artifacts
    3. Additive T1 and T2 contrast
    4. Lower specific absorption rate (SAR)

    In conventional SE imaging combined T1 and T2 effects are generally competitive, meaning that ↑T1 generally results in decreased signal intensity while ↑T2 results in increased intensity. In IR imaging by the appropriate choice of TR, T1 and T2 effects can be separated and made to be additive. Link to Q&A discussion

  9. On a magnitude-reconstructed IR image obtained at 1.5 T, which inversion time would be useful for suppressing fat?
    1. TI = 170
    2. TI = 450
    3. TI = 980
    4. TI = 2000

    For most tissues the null point typically occurs when TI ≈ 0.69 x T1. For fat with a T1 value at 1.5 T of approximately 250, this corresponds to a TI of approximately 170 ms. Link to Q&A discussion

  10. Which of the following statements concerning phase-sensitive inversion recovery (PSIR) is false?
    1. Short T1 tissues always have a signal brighter than long T1 tissues, regardless of TI.
    2. Background air always appears black.
    3. Selective nulling of tissues by adjusting TI is not possible as it is for magnitude-reconstructed IR.
    4. It is useful for assessing pediatric myelination and myocardial contrast enhancement.

    Phase-sensitive IR preserves the information about the polarity of the longitudinal magnetization after the 180°(inverting)-pulse, with more negative values rendered increasingly darker. Background air is generally rendered a middle shade of gray, not black. Short T1 tissues are always brighter than long T1 tissues, but both can be darker than air at short TI’s. Common applications are in pediatric and cardiac imaging. Link to Q&A discussion

  11. Concerning the null point value (TInull) for a given tissue in magnitude-reconstructed IR, which statement is false?
    1. TInull increases with increasing field strength.
    2. TInull increases with increasing T1 value
    3. For long TR, TInull ≈ 0.69 x T2.
    4. TInull depends on whether the signal is generated by a conventional spin-echo or fast spin-echo sequence.

    For TR>>T1, TInull = T1 • (ln 2) ≈ 0.69 x T1, (not 0.69 x T2, so answer c is false). Because tissue T1 increases with field strength, TInull does also. TInull depends on echo time of the last echo in a FSE sequence. Link to Q&A discussion

  12. Which of the following statements about the Short TI Inversion Recovery (STIR) sequence is false?
    1. It is a very useful technique for fat saturation.
    2. It is a very useful technique for detecting gadolinium enhancement.
    3. It has additive T1 and T2 effects.
    4. It is especially useful in lower field permanent magnets with relatively poor homogeneity.

    STIR cannot be used as a fat suppression technique post-gadolinium. STIR does not specifically suppress fat; it only suppresses tissues with T1 values in the range of fat (200-300 ms), so gadolinium enhancement would be suppressed. Link to Q&A discussion

  13. Which of the following pulse sequence parameters might be used to create a T1-FLAIR brain image?
    1. TR = 2500, TE = 10, TI = 900
    2. TR = 2500, TE = 50, TI = 180
    3. TR = 4000, TE = 80, TI = 1800
    4. TR = 8000, TE = 100, TI = 2500

    Answer (a) is correct. Answer (b) would be considered a STIR image, while (c) and (d) would be T2-FLAIR images. Link to Q&A discussion

  14. Double Inversion Recovery (DIR) sequences for brain imaging utilize two different TI values, which typically suppress
    1. Fat and CSF
    2. CSF and white matter
    3. CSF and gray matter
    4. Gray matter and white matter

    DIR brain imaging is particularly useful for detecting lesions of the white matter and cortex. Typically the first 180°-pulse suppresses CSF and the second suppresses white matter. Link to Q&A discussion

Back to top of page

Gradient Echo and Susceptibility

  1. Which of the following is not an advantage of gradient echo imaging over spin-echo imaging?
    1. Shorter TEs
    2. Shorter TRs
    3. Reduced flow artifacts
    4. Reduced susceptibility artifacts

    Phase shifts resulting from magnetic field inhomogeneities, static tissue susceptibility gradients, or chemical shifts are not cancelled at the center of the GRE as they are in SE sequences. Image contrast is therefore dictated not by true T2-relaxation, but by these other factors which constitute T2*. GRE sequences are therefore more frequently troubled by susceptibility and chemical shift artifacts and do not function well on scanners whose magnetic fields lack homogeneity Link to Q&A discussion

  2. Concerning multi-echo GRE, which statement is true?
    1. The peak value of successive echoes is determined by the T2 decay of the tissue imaged.
    2. The peak value of successive echoes echo is determined by the T1 value of the tissue imaged.
    3. One common application is for fat-water imaging.
    4. Each successive echo must be preceded by its own RF-pulse.

    The peak value of successive echoes is determined by the T2* decay (not T2 decay) of the tissue imaged (options a and b are false). One common application of multi-echo GRE is for acquiring fat-water "in-phase" and "out-of-phase" images using two different TE's (options c is true). Only one RF pulse is used for the entire chain of echoes (option d is false). Link to Q&A discussion

  3. Which is not a technique for gradient echo spoiling?
    1. Use of very short TR
    2. Use of 2D multislice mode
    3. Use of a variable spoiler gradient
    4. Semirandom phase change of the RF-carrier wave

    All are valid spoiling methods except choice (a). Long TR values produce “natural” spoiling, not short TRs. When TR>>T2*, the transverse magnetization will naturally decay to zero by the end of the cycle. Thus any gradient echo sequence using TR values of several hundred milliseconds will no longer have appreciable transverse coherences and be “spoiled”. Link to Q&A discussion

  4. A spoiled GRE sequence with TR = 10, TE = 3, and α = 50º would be considered primarily
    1. T1-weighted
    2. T2-weighted
    3. T2*-weighted
    4. Proton-density-weighted

    Short TR and large α accentuate T1-weighted imaging. Link to Q&A discussion

  5. A spoiled GRE sequence with TR = 1000, TE = 30, and α = 5º would be considered primarily
    1. T1-weighted
    2. T2-weighted
    3. T2*-weighted
    4. Proton-density-weighted

    Long TR and small α minimize T1 weighting, while long TE accentuates T2* weighting. spoiled-gre-parameters.html Link to Q&A discussion

  6. A spoiled GRE sequence with TR = 1000, TE = 2, and α = 5º would be considered primarily
    1. T1-weighted
    2. T2-weighted
    3. T2*-weighted
    4. Proton-density-weighted

    Long TR and small α minimize T1 weighting, while short TE minimizes T2* weighting. This leaves the image more proton-density weighted than the other examples. Link to Q&A discussion

  7. In a spoiled GRE sequence, a low flip angle (α)
    1. Minimizes T1 weighting
    2. Accentuates T1 weighting
    3. Minimizes T2* weighting
    4. Minimizes proton density weighting

    Flip angle (α) controls T1-weighting. A small flip angle minimizes T1-weighting because the longitudinal magnetizations of various tissues are not differentiated much by such a small angular displacement. Hence at small flip angles, [H] and T2* effects predominate. Conversely, as α → 90º, T1-weighting increases. Link to Q&A discussion

  8. In a spoiled GRE sequence, a long TE
    1. Maximizes T2 weighting
    2. Maximizes T2* weighting
    3. Minimizes T2 weighting
    4. Minimizes T2* weighting

    T2*-weighting increases as TE is prolonged. This is because a longer TE allows more time for dephasing before echo formation. Link to Q&A discussion

  9. Which of the following SSFP pulse sequences is considered “balanced”?
    1. TrueFISP/FIESTA
    2. FISP/GRASS/FFE
    3. DESS/MENSA
    4. PSIF/T2-FFE

    Sequences such as TrueFISP (Siemens) and FIESTA (GE) sample both FID-like and Echo-like SSFP signals and are considered “balanced”. Link to Q&A discussion

  10. Which of the following is not a condition for a SSFP signal to exist?
    1. TR must be longer than T2.
    2. Phase shifts caused by gradients must be constant from cycle to cycle.
    3. Field inhomogeneities must be static.
    4. Spins must be stationary or motion-compensated.

    Choice (a) is incorrect. TR must be significantly shorter (not longer) than T2, or else natural decay processes will destroy the transverse coherence. Link to Q&A discussion

  11. Concerning “coherent” gradient echo sequences like GRASS/FISP, which statement is false?
    1. If TR is long compared to T2*, the contrast resembles that of a spoiled GRE sequence.
    2. TE controls T2* weighting.
    3. Small flip angles (α) produce T1-weighted images.
    4. Large flip angles produce images that are weighted by T2/T1.

    Choice (c) is incorrect. A small flip angle minimizes T1-weighting because the longitudinal magnetizations of various tissues are not differentiated much by such a small angular displacement. Hence at small flip angles, [H] and T2* effects predominate. Link to Q&A discussion

  12. An important tissue weighting present in balanced SSFP sequences like True FISP and FIESTA is
    1. T1
    2. T2*
    3. T1 + T2
    4. T2/T1

    True FISP sequences behave more like spin echo than gradient echo sequences in that they do not have T2*-dependence. Also, since TR is nearly always much, much shorter than T1 or T2, the signal intensity is related to the ratio T2/T1. Link to Q&A discussion

  13. Excluding fluids (like urine, blood, and CSF), which of the following tissues is the brightest on balanced SSFP sequences like True FISP and FIESTA?
    1. Fat
    2. Myocardium
    3. Liver
    4. Brain

    Fat is the second brightest tissue on balanced SSFP images after pure fluids. Link to Q&A discussion

  14. What is the explanation for the brightness of fluid and the tissue identified in the prior question?
    1. The T2 values are long.
    2. The T2* values are long.
    3. The T2/T1 ratios are large.
    4. The T1+T2 values are large.

    Balanced SSFP sequences have the special property that their contrast is dependent not on just T1 or T2, but the ratio T2/T1. For most solid tissues, T1>>T2 so the T2/T1 ratio is small (< 0.10). The exceptions are fat (T2/T1 ≈ 0.30, due to its short T1) and fluids (T2/T1 ≈ 0.70, due to the fact that T1 and T2 are approximately equal). Link to Q&A discussion

  15. TEXTFORquestion15
    1. There is no substantial difference; just name changes for marketing purposes.
    2. More phase banding artifacts are present in CISS and FIESTA-C.
    3. CISS/FIESTA-C images are T1-weighted
    4. CISS/FIESTA-C consist of two balanced SSFP sequences run back-to-back.

    FIESTA-C/CISS is composed of a pair of TrueFISP acquisitions run back-to-back preceded by an automatic shimming procedure. The first uses phase alternation of the RF-pulses (+α, −α, +α, −α, ...) while the second does not (+α, +α, +α, etc). When the paired data sets are combined in maximum intensity projection, the phase errors cancel, resulting in an image largely free of dispersion banding. Link to Q&A discussion

  16. Which of the following statements about DESS (dual-echo steady state) is false?
    1. Its main use is in musculoskeletal imaging.
    2. It is insensitive to motion.
    3. FID-like and Echo-like signal components are recorded separately during a single TR interval.
    4. Fluids are very bright while trabecular bone is very dark.

    By means of a prolonged and unbalanced readout gradient, DESS generates the FID-like and Echo-like signals from the steady-state free precession individually. It then combines the signals on a pixel-by-pixel basis. DESS is very sensitive to motion, so option (b) is false. Its main use is in musculoskeletal imaging where joint fluid is bright and bone is dark. Link to Q&A discussion

  17. Which of the following statements about the GRASE (Gradient and Spin Echo) pulse sequence is true?
    1. It is a widely used sequence for routine head imaging.
    2. Its tissue energy deposition (specific absorption rate) is higher than with conventional spin echo sequences.
    3. It is superior to conventional spin echo (CSE) or fast spin echo (FSE) for detecting calcifications and hemorrhages.
    4. It is primarily used to generate T1-weighted images.

    GRASE is a hybrid technique supposedly combines the best features of CSE and GRE imaging. The GRE contribution makes it useful for detecting calcification of blood (although less so than a pure GRE sequence). Its SAR is lower than a comparable fast spin-echo sequence because there are fewer RF-pulses. GRASE has never achieved much popularity as an imaging method. Link to Q&A discussion

  18. The SI unit for magnetic susceptibility (χ)
    1. Tesla
    2. Oersted
    3. Ampere-meter
    4. Dimensionless

    Magnetic susceptibility, denoted by the Greek letter chi (χ), is defined as the magnitude of the internal polarization (J) divided by the strength of the external field (B). Since it is the ratio of two magnetic fields, susceptibility is a dimensionless number. Link to Q&A discussion

  19. Concerning diamagnetism, which statement is false?
    1. The internal magnetization/polarization of a diamagnetic substance opposes the applied field.
    2. Diamagnetic substances have positive susceptibilities (χ > 0).
    3. Water, and hence most tissues, are diamagnetic.
    4. Calcifications are diamagnetic.

    Diamagnetic substances become polarized in a manner that opposes the applied magnetic field. They have negative, not positive, susceptibilities (option b is false). Water, calcifications, and most tissues are diamagnetic. Link to Q&A discussion

  20. Concerning paramagnetism, which statement is false?
    1. The internal magnetization/polarization of a paramagnetic substance is in the same direction as the applied field.
    2. Air is paramagnetic.
    3. Gadolinium contrast material is paramagnetic.
    4. Most steels are paramagnetic.

    Paramagnetic substances become polarized in a manner that augments the applied magnetic field. They have positive susceptibilities. Surprisingly, molecular oxygen is mildly paramagnetic, so air is as well. Gadolinium contrast is paramagnetic. Most steels are ferromagnetic, a phenomenon resembling, but many orders of magnitude greater than paramagnetism. Link to Q&A discussion

  21. Which of the following 2D gradient echo sequences would be most useful for demonstrating small calcifications or hemorrhages in the brain?
    1. TR = 800, TE = 30, α = 20°
    2. TR = 2000, TE = 10, α = 70°
    3. TR = 60, TE = 3, α = 45°
    4. TR = 800, TE = 10, α = 45°

    Hemorrhage/calcification GRE sequences are characteristically operated in 2D multi-slice mode using relatively long TR's and low flip angles (both minimizing T1 effects) and relatively long TE's (to accentuate T2* dependence). So choice (a) is the best answer. Link to Q&A discussion

  22. Concerning susceptibility-weighted imaging (SWI), which of the following is false?
    1. SWI images are typically acquired in 3D rather than 2D mode.
    2. Magnitude and phase information can be individually viewed or combined.
    3. SWI imaging is inferior to 3D GRE studies for detecting microhemorrhages and calcifications.
    4. SWI studies typically include minimum intensity images.

    Multiple studies have shown that SWI imaging is superior to 3D GRE studies for detecting microhemorrhages and calcifications. Link to Q&A discussion

  23. Concerning SWI phase map images, which statement is incorrect?
    1. A lesion that is bright on a GE scanner will appear dark on a Siemens scanner.
    2. Phase map images are displayed using a minimum intensity projection algorithm.
    3. If a venous sinus appears dark on a phase image, paramagnetic blood products will also be dark.
    4. The choroid plexus and pineal gland provide a good internal reference for diamagnetic substances.

    Phase map images are rendered in a simple magnitude mode on a slice-by-slice basis, so option (b) is false. The "colors" (black or white) of diamagnetic and paramagnetic substances on SWI phase images are scanner- dependent, so reference to a known tissue (veins for paramagnetic, choroid plexus calcification for diamagnetic) are needed as an internal check. Link to Q&A discussion

Back to top of page

Diffusion

  1. The units for the diffusion coefficient (D) are
    1. mm/sec
    2. mm²/sec
    3. mm³/sec
    4. Dimensionless

    D reflects the flux of particles through a surface during a certain period of time, it therefore has units of area/time (e.g. mm²/sec). Link to Q&A discussion

  2. According to the Stokes-Einstein equation, which of the following system changes would increase the diffusion coefficient? (You should be able to reason this out even if you’ve forgotten the exact equation).
    1. Increasing temperature
    2. Increasing viscosity of the medium
    3. Increasing the size of the particles
    4. All of the above
    5. None of the above

    Increasing temperature increases the kinetic energy and hence motion of the particles, so D is increased. Bigger particles and ones in more viscous media move more slowly, so D is decreased. Link to Q&A discussion

  3. Comparing the diffusion coefficients of water (Dw) and biological tissues (Dt), which statement is most accurate?
    1. Dw < Dt
    2. Dw and Dt are almost exactly the same
    3. Dt is typically only 10-50% of Dw
    4. Dt is typically less that 1% of Dw

    The diffusion constant for pure water at body temperatures is approximately 3.0 x 10−3 mm2/sec. The values of D for biological tissues are only 10-50% as long, perhaps 1.0 x 10−3mm2/sec on the average. Link to Q&A discussion

  4. How many elements are there in a first-order diffusion tensor for describing anisotropic materials?
    1. 3
    2. 6
    3. 9
    4. 12

    In anisotropic materials, diffusion cannot be described by a single number, but a [3 x 3] array called the diffusion tensor. The three diagonal elements (Dxx, Dyy, and (Dzz) of the tensor represent diffusion coefficients measured in the laboratory frame of reference along each of the principal (x-, y- and z-) directions. The six off-diagonal terms (Dxy, Dyz, etc) reflect correlation between random motions corresponding to each pair of principal directions. Link to Q&A discussion

  5. Modern diffusion-weighted pulse sequences all trace their origin to the pulsed gradient spin echo (PGSE) technique developed by
    1. Stejskal and Tanner
    2. Solomon and Blombergen
    3. Carr and Purcell
    4. Einstein and Stokes

    The pulsed gradient spin echo (PGSE) technique was developed by chemists Edward Stejskal and John Tanner in the mid-1960's. It consisted of paired diffusion sensitizing gradients flanking either side of a 180º-inversion pulse. Modern variations include addition of a chemically selective fat suppression pulses, use of bipolar gradients, and a second 180º pulse immediately before the image acquisition module. Link to Q&A discussion

  6. How can a larger b-value be achieved with paired pulsed diffusion gradients?
    1. By increasing the amplitude of the gradients
    2. By increasing the duration of the gradients
    3. By widening the time interval between the two gradients
    4. All of the above
    5. None of the above

    The b-value depends on the strength, duration, and spacing of the pulsed diffusion gradients. A larger b-value is achieved with increasing the gradient amplitude and duration and by widening the interval between gradient pulses. Link to Q&A discussion

  7. What is the approximate range of b-values used to produce diffusion-weighted images in standard clinical MRI today?
    1. 0 – 10 s/mm²
    2. 0 – 100 s/mm²
    3. 0 – 1,000 s/mm²
    4. 0 – 10,000 s/mm²

    Most routine clinical DWI currently use b-values between 0 and 1000 or perhaps 0 and 2000, with slightly lower values being used outside the central nervous system. Link to Q&A discussion

  8. What are the minimum number of source image gradient directions that must be applied to obtain a diffusion-weighted anatomic image?
    1. 1
    2. 3
    3. 9
    4. 27

    At least three sets of source images must be obtained. These may be along the laboratory x-, y-, and z-axes or in three arbitrary perpendicular orientations. More modern schemes typically obtain source images in 6, 20 or more directions, but three is the minimum. Link to Q&A discussion

  9. What is the definition of the “trace” of a 3x3 diffusion tensor matrix?
    1. The largest element
    2. The sum of all the elements
    3. The sum of the diagonal elements
    4. The average of the diagonal elements

    The term "trace" comes from matrix algebra where it means the sum of diagonal elements of such an array. The trace of the diffusion tensor equals (Dxx + Dyy + Dzz). Link to Q&A discussion

  10. If the trace of a 3x3 diffusion tensor matrix corresponding to a single pixel is calculated to be 1.5 x 10–3 s/mm², what is the corresponding apparent diffusion coefficient (ADC)?
    1. 0.5 x 10–3 s/mm²
    2. 1.5 x 10–3 s/mm²
    3. 3.0 x 10–3 s/mm²
    4. 4.5 x 10–3 s/mm²

    The apparent diffusion coefficient (ADC) is considered to be the average value of the trace, or ADC = (Dxx + Dyy + Dzz) / 3. So the correct answer is (a). Link to Q&A discussion

  11. Concerning the ADC map, which one of the following statements is true?
    1. A lesion bright on a trace DWI image will be dark on an ADC map.
    2. Lesions with very long T2 values will appear dark on an ADC map.
    3. Normal urine will appear dark on a trace DWI image and bright on an ADC map.
    4. The ADC map is simply the inverse of the trace DWI image.

    The Trace DW image is not a map of diffusion; it is only diffusion-weighted, a fact implicit in its name. Trace DW images possess considerable T2-weighting. As such, lesions with either very long or very short T2 values may "contaminate" the Trace DW images, making them appear "artificially" bright or dark. These important phenomena are known as "T2-shine-through" and "T2-black-out. T2-effects can be mathematically removed from the DW image to create a pure parametric image of apparent diffusion coefficients (the "ADC Map"). The ADC map is not a simple inverse of the DWI trace image. Link to Q&A discussion

  12. Which of the following is incorrect concerning exponential ADC maps?
    1. eADC maps have a gray scale that parallels that seen in the trace DW images.
    2. eADC maps are simply the trace DW image divided by the b0 image.
    3. eADC maps have more noise than trace DW images.
    4. eADC maps still have the problem of “T2-shine-through”.

    The eADC map is simply the trace DW image divided by the b0 image for each point. Because T2 effects in the numerator and denominator cancel, eADC maps eliminate the problem of T2-shine-through. (Option d is false.) The signal intensities are more intuitive because they follow that seen on the trace images. However, because the exponential image is a calculated map, like the ADC map it is often of lower quality and noisier than the DW image. Link to Q&A discussion

  13. A cerebral hematoma 36-hours-old appears dark on both the b0 and trace DW images. What is the explanation?
    1. Diffusion is not restricted in the hematoma.
    2. This is an example of “T2-shine-through”.
    3. This is an example of “T2-blackout”.
    4. This is an example of diffusion anisotropy.

    This case illustrates the “T2-blackout” phenomenon, where severe T2* shortening (here due to paramagnetic deoxyhemoglobin in the hematoma) spills over and “contaminates” the trace DW images. Diffusion is actually restricted in the hematoma (think of all the clumped red blood cells and fibrin), but appears (falsely) dark on the DW image. Link to Q&A discussion

  14. Which one of the following mechanisms does not explain restricted diffusion in an acute cerebral infarction?
    1. Increased intracellular water (cytotoxic edema)
    2. Decreased intracellular and extracellular viscosity.
    3. Reduction in extracellular space.
    4. Fragmentation of cellular components.

    All are correct except for (b). As cellular components fragment and proteins unravel, viscosity increases (not decreases). Increased viscosity restricts diffusion because it inhibits the movement of water molecules through the diseased tissue. Link to Q&A discussion

  15. Which of the following tumors would not be expected to demonstrate restricted diffusion?
    1. Splenic lymphoma
    2. Cerebellar medulloblastoma
    3. Pancreatic serous cystadenoma
    4. Ewing’s sarcoma (PNET) of the pelvis

    Highly cellular neoplasms with densely packed cells and relatively little extracellular space often demonstrate restricted diffusion. Examples include lymphomas, medulloblastomas, primitive neuroectodermal tumors (PNETs) and highly malignant gliomas. A cystic tumor (like option c would not be expected to restrict diffusion. Link to Q&A discussion

  16. Assuming mirror-image off-diagonal elements are equal, how many unique elements does a first-order [3x3] diffusion tensor contain?
    1. 3
    2. 6
    3. 9
    4. 12

    The off-diagonal elements being equal implies that diffusivities should be the same in the forward and reverse directions (i.e., x→y and y→x diffusivities are the same). This means the tensor is symmetric and 3 of the 9 tensor elements are equal (Dxy = Dyx, Dxz = Dzx, and Dyz = Dzy). This leaves 6 unique elements to be measured. Link to Q&A discussion

  17. How many measurements does it take to determine the unique diffusion tensor elements in the prior question?
    1. 5
    2. 6
    3. 7
    4. 9

    Six measurements need to be taken with the diffusion sensitizing gradients applied, plus one b0 measurement, giving a total of 7. Link to Q&A discussion

  18. Concerning the diffusion ellipsoid, which statement is false?
    1. The directions of its major and minor axes are described by eigenvectors.
    2. The radii of its major and minor axes are eigenvalues.
    3. The eigenvalues are proportional to Einstein’s root mean square diffusion displacement in each direction.
    4. The diffusion tensor matrix for the diffusion ellipsoid has 6 unique elements.

    A significant benefit to using the diffusion ellipsoid is that in this frame of reference, the off-diagonal elements of the diffusion tensor disappear. The set of eigenvalues define a matrix with only unique 3 diagonal elements (λ1, λ2, and λ3) with the off-diagonal elements all zeroes. Link to Q&A discussion

  19. Concerning fractional anisotropy (FA), which statement is incorrect?
    1. FA varies from −1 to +1.
    2. FA is an index for the amount of diffusion asymmetry within a voxel.
    3. FA = 0 for a voxel with perfect isotropic diffusion.
    4. On an FA map, brighter areas are more anisotropic than darker areas.

    The value of FA varies between 0 and +1 (option a is incorrect). For perfect isotropic diffusion, λ1 = λ2 = λ3, the diffusion ellipsoid is a sphere, and FA = 0. With progressive diffusion anisotropy, the eigenvalues become more unequal, the ellipsoid becomes more elongated, and the FA → 1. The FA map is a gray-scale display of FA values across the image. Brighter areas are more anisotropic than darker areas. Link to Q&A discussion

  20. Concerning whole-body DWI, which statement is false?
    1. Imaging is obtained in multiple blocks or stations, then digitally stitched together.
    2. Views of the chest and abdomen must be obtained during breath holding to reduce motion artifact.
    3. Standard image display is in a black-background mode to resemble PET-CT.
    4. Each DWI sequence is preceded by a STIR-like inversion pulse for fat suppression.

    Whole-body DWI does not require breath holding. Even though the organs of the abdomen and chest move during image acquisition, they do so "coherently". Their physical displacements are cyclic and while this motion produces some spatial blurring it does not significantly affect the magnitude of the DW signal. Link to Q&A discussion

  21. All of the following are advantages of readout-segmented DWI over single-shot DWI, except
    1. Increased signal-to-noise
    2. Decreased imaging time
    3. Decreased susceptibility artifacts
    4. Reduced spatial blurring

    Depending on the number of “shots” readout segmented DWI can take 3-5 times longer than single-shot DWI, so option (b) is false. Link to Q&A discussion

  22. The key technological development underlying modern small field-of-view DWI methods (like ZOOMit and FOCUS) is
    1. Use of 2D spatially-selective composite RF pulses
    2. Use of outer volume saturation pulses
    3. Use of stimulated echo inner volume pulses
    4. Use of extremely strong, continuously applied gradients

    Newer small FOV DWI sequences employ a special 2D RF excitation pulse that is spatially selective in both the slice select and phase-encoding directions. The 2D RF "pulse" is a composite of approximately 25 closely spaced "sub-pulses" extending over a time frame of about 15 ms. This RF excitation is played out simultaneously with a fast (oscillatory) gradient along the phase-encode axis and a slow ("blipped") gradient along slice-select. Link to Q&A discussion

  23. Which of the following statements about intravoxel incoherent motion (IVIM) is true?
    1. IVIM effects are most easily recognized when high b-values are used.
    2. IVIM describes signal losses due to both diffusion and microscopic perfusion.
    3. IVIM perfusion effects mitigate the signal losses caused by diffusion alone.
    4. Le Bihan’s IVIM model allows estimation of capillary blood flow.

    The correct statement is (b): IVIM describes signal losses due to both diffusion and microscopic perfusion. If the gradients are relatively strong, IVIM-induced signal losses are primarily due to diffusion — the Brownian motion of water molecules in and around cells. When weaker gradients are used, however, a second IVIM mechanism also contributes to signal loss — microcirculation of blood in the capillary network. Le Bihan’s IVIM model allows estimation of the perfusion fraction (the percent of a voxel volume occupied by capillaries), but not the blood flow through them. Link to Q&A discussion

  24. Which of the following statements about diffusion kurtosis is false?
    1. Diffusion kurtosis measures the non-Gaussian movement of water molecules.
    2. Kurtosis effects are more noticeable when long echo times are used.
    3. Kurtosis effects are more noticeable when low b-values are used.
    4. The mean kurtosis (K) of a pure fluid is zero.

    Standard diffusion weighted imaging (DWI) methods have incorporated Einstein's original concept that the diffusion water molecules follows a Gaussian (normal) distribution. Non-Gaussian behavior becomes more noticeable when stronger gradients (higher b-values) and longer echo times are used. By definition, a Gaussian distribution has K = 0, which would be the case with a pure fluid. Link to Q&A discussion

Back to top of page

Fat-Water

  1. What is the main type of fat within the body which contributes most to the signal recorded on MRI?
    1. Triglycerides
    2. Free fatty acids
    3. Cholesterol
    4. Phospholipids

    The bulk of the MR signal in fatty human tissues arises from triglycerides, with lesser contributions from free fatty acids and cholesterol (when esterified or in a semi-solid or liquid state). Link to Q&A discussion

  2. At a given field strength, how do the nuclear precession frequencies of ¹H protons differ between fat and water?
    1. Fat and water protons precess at the same frequency
    2. Fat protons precess faster than water protons.
    3. Fat protons precess slower than water protons.
    4. Fat protons may precess either faster or slower depending on their location along the triglyceride side chains.

    The H protons of triglycerides are nestled within long carbon chains, shielded by electron clouds and protected from the full force of the externally applied magnetic field. In water, by comparison, the electronegative O atom pulls protective electron clouds away from the H nuclei, exposing them to the full force of the external magnetic field. Thus water protons resonate faster than fat protons. Link to Q&A discussion

  3. A chemical shift of 3.5 ppm written in decimal form is
    1. 0.00000035
    2. 0.0000035
    3. 0.000035
    4. 0.00035

    The abbreviation ppm stands for “parts per million” and is equivalent to multiplying the base value by 10−6. So 3.5 ppm = 3.5 x 10−6 = 0.0000035 (answer b). Link to Q&A discussion

  4. If the water-fat chemical shift is 3.5 ppm, calculate the fat-water frequency difference for a 7T scanner operating at 200 MHz.
    1. 220 Hz
    2. 440 Hz
    3. 700 Hz
    4. 900 Hz

    Δf = (200 MHz)(3.5 ppm) = (200 x 106 Hz)(3.5 x10−6) ≈ 700 Hz Link to Q&A discussion

  5. If the water-fat chemical shift is 3.5 ppm, calculate the water-fat frequency difference for a portable 0.064T scanner operating at 150 kHz.
    1. 0.5 Hz
    2. 5 Hz
    3. 50 Hz
    4. 500 Hz

    Δf = (150 kHz)(3.5 ppm) = (150 x 103 Hz)(3.5 x10−6) ≈ 0.5 Hz. This very low frequency separation means that water-fat artifacts will be minimal at this field strength, and also that spectral fat suppression methods cannot be used. Link to Q&A discussion

  6. At 1.5 T water and fat go in and out of phase about every 2.2 msec. How fast would they go in and out of phase at 3.0T?
    1. Every 1.1 msec
    2. Every 2.2 msec
    3. Every 3.3 msec
    4. Every 4.4 msec

    The period of this phase cycling is 1/Δf, where Δf is the frequency offset between the spins. So if the field is doubled, the frequency offset is twice as large, and the phase cycling period is half as large, so (a) 1.1 msec is the correct answer. Link to Q&A discussion

  7. An adrenal mass that is of intermediate signal intensity on an in-phase image and low signal on an out-of-phase image is most likely an
    1. Adrenal carcinoma
    2. Adrenal adenoma
    3. Adrenal metastasis
    4. Adrenal pheochromocytoma

    Adrenal adenomas, the most common tumor of the adrenal gland, often contain microscopic lipid droplets. Voxels containing both lipid and water result in signal cancellation with low signal on out-of-phase GRE images. Link to Q&A discussion

  8. Which of the following is not one of the four standard images produced by most commercial Dixon-type sequences?
    1. Chemical shift
    2. Water
    3. Fat
    4. In-phase
    5. Out-of-phase

    Choice (a), chemical shift, is not a standard image generated by a Dixon sequence. The other four are. Link to Q&A discussion

  9. Concerning CHESS/Fat-Sat pulses, which one of the following statements is incorrect?
    1. They reduce the number of available slices for a given TR.
    2. They cause tissue heating.
    3. They include a spectrally tuned RF-pulse followed by spoiler gradients along one or more axes.
    4. They are the preferred method for fat suppression at 0.3T and below.

    All are correct except choice (d). The effectiveness of a Fat-Sat pulse depends primarily on field strength and field homogeneity. At fields below 0.3T the water and fat peaks are so close together (in Hz) that it is difficult to cleanly suppress one or the other with a chemically selective pulse. At low fields another method of fat suppression (typically STIR or Dixon) must therefore be used. Link to Q&A discussion

  10. Which of the following flip angle patterns would not be considered to define a composite binomial pulse for water excitation?
    1. 1:1
    2. 1:1:1
    3. 1:2:1
    4. 1:3:3:1

    Binomial pulses are typically used for selective water excitation as a means of fat suppression. Binomial pulses have flip angles that follow the pattern of coefficients of the binomial expansion of (a+b)n: 1-1, 1-2-1, 1-3-3-1, etc. Thus, a 90º-pulse could be constructed as a [45º-45º] pair, a [22.5º-45º-22.5º] triplet, or a [11.25º-33.75º-33.75º-11.25º] quadruplet. So choice (b), 1:1:1 is not a binomial pulse. Link to Q&A discussion

  11. Limitations of the STIR technique for fat suppression include all except
    1. Inability to use it to detect gadolinium enhancement.
    2. Suppression of other short T1 materials besides fat (eg, protein, blood).
    3. Tissue heating from multiple 180º pulses.
    4. Decreased visualization of long T1/long T2 lesions due to competitive signal effects.

    All are valid limitations except for (d), which is incorrect. In routine SE imaging lesions with prolonged T1 and T2 have competitive effects on signal intensity (↑T1 reduces signal while ↑T2 increases signal). In STIR imaging the effects of ↑T1 and ↑T2 are additive, allowing for improved visualization of some lesions, such as multiple sclerosis plaques. Link to Q&A discussion

  12. The Spectral Presaturation with Inversion Recovery (SPIR) technique can be thought of as a hybrid combining fat suppression features of
    1. Spectral Fat Sat (CHESS) + STIR
    2. Dixon Technique + STIR
    3. Spectral Fat Sat (CHESS) + Selective Water excitation
    4. Dixon Technique + Selective Water excitation

    SPIR is a hybrid technique that combines a fat-selective RF-pulse and spoiler gradient (similar to CHESS) together with nulling of the residual longitudinal fat magnetization through an inversion delay mechanism (similar to STIR). These spin manipulations purely involve fat; the water resonance is unaffected. After a suitable inversion time to null residual fat signal, any pulse sequence can be used to image the remaining water. Link to Q&A discussion

  13. Which of the following statements comparing SPAIR (SPectral Attenuated Inversion Recovery) and SPIR (Spectral Presaturation with Inversion Recovery) is incorrect?
    1. SPAIR uses adiabatic pulses while SPIR does not.
    2. SPAIR uses an inversion pulse of 180º, while SPIR uses pulses in the 100º -120º range
    3. SPAIR deposits less RF energy into tissue than does SPIR.
    4. The inversion time is longer for SPAIR than SPIR.

    Both SPAIR and SPIR can be thought of as CHESS-STIR hybrids. SPAIR uses 180º adiabatic pulses while SPIR uses non-adiabatic pulses in the 100º -120º range. Accordingly-SPAIR pulses deposit more RF-energy in tissue than the smaller flip angle SPIR or CHESS pulses. The inversion time is longer in SPAIR than SPIR, so there is a greater penalty in terms of imaging time and reduced number of slices for a given TR. Link to Q&A discussion

  14. Which of the following statements about adiabatic RF-pulses is false?
    1. They are both amplitude- and frequency-modulated.
    2. Their transmitted frequency changes simultaneously with amplitude as the pulse evolves.
    3. They are less sensitive to B1 inhomogeneities that other pulses.
    4. They are commonly used in low-field applications.

    Only (d) is false. Like other non-adiabatic but spectrally selective pulses (CHESS, STIR) they can only be used at higher, rather homogeneous fields where there is a clean separation of water and fat resonances Link to Q&A discussion

Back to top of page