Pulmonary Magnetic Resonance Imaging

Chapter 5

MRI

Abstract

ways where <sup>3</sup>

Keywords: X-Centric, <sup>3</sup>

hyperpolarized propane (<sup>1</sup>

51

Our main focus will be on <sup>3</sup>

bronchodilation [22, 23]. Small animal <sup>3</sup>

1. Introduction

High Resolution <sup>3</sup>

Matthew S. Fox and Alexei V. Ouriadov

development of new therapies. Small animal <sup>3</sup>

called X-Centric is described for acquiring <sup>3</sup>

imaging techniques such as hyperpolarized <sup>3</sup>

gas (19F) MRI [17–19] to better understand various lung diseases.

Hyperpolarized gas MRI of the mouse lung is of great interest due to the urgent need for novel biomarkers for the assessment of respiratory-disease progression and

He diffusion is maximized. In this chapter, a modified FGRE approach

He, 129X, 19F mouse, rat, lung, short-TE, x-nuclei,

resolution imaging (<500 μm) to obtain acceptable images for visualization of all branches of lung microstructure from the mouse trachea to lung parenchyma. The use of conventional fast-gradient-recalled echo (FGRE) pulse sequences for highspatial-resolution mouse lung imaging is challenging due to the signal loss caused by significant diffusion-weighting by the imaging gradients, particularly in larger air-

center-out technique, allowing high-spatial-resolution, and high signal-to-noise ratio density-weighted imaging, as it is a short-TE method minimizing diffusion decay. Here, we also take advantage of a high-performance insertable-gradient-set interfaced with a clinical MRI system and a custom-built constant-volume ventila-

Centric imaging was performed in a phantom and mouse lungs, yielding a nominal resolution of 39 μm and 78 μm respectively. We also demonstrate the feasibility of

pre-clinical, hyperpolarized, MRI, COPD, CF, BPD, alpha-1 antitrypsin deficiency

Inhaled hyperpolarized gas lung MRI [1] was proven to be useful for the observation and treatment planning of several pulmonary diseases including chronic obstructive pulmonary disease (COPD) [2–6], asthma, [7–11] and lung cancer [12]. With the combined economic burden of COPD and asthma in Canada, Ontario being \$5.7 billion (2011) [13], and lung cancer being the leading cause of cancer deaths worldwide, [14] there has been a growing interest in developing new lung

[20, 21], as it is a promising tool to quantify airway and ventilation abnormalities

associated with bronchoconstriction, airway narrowing, and subsequent

tor to get the maximum benefits of X-Centric. High-spatial-resolution <sup>3</sup>

129Xe/19F X-Centric MRI in a phantom and in rat lungs.

He Pulmonary

He lung MRI requires high-spatial-

He mouse lung MRI. X-Centric is a

He and 129Xe MRI as well as

He lung MRI normally assumes high spatial

H) MRI [15, 16] and thermally polarized inert fluorinated

He gas MRI of small animal models of lung disease

He X-

#### Chapter 5

## High Resolution <sup>3</sup> He Pulmonary MRI

Matthew S. Fox and Alexei V. Ouriadov

#### Abstract

Hyperpolarized gas MRI of the mouse lung is of great interest due to the urgent need for novel biomarkers for the assessment of respiratory-disease progression and development of new therapies. Small animal <sup>3</sup> He lung MRI requires high-spatialresolution imaging (<500 μm) to obtain acceptable images for visualization of all branches of lung microstructure from the mouse trachea to lung parenchyma. The use of conventional fast-gradient-recalled echo (FGRE) pulse sequences for highspatial-resolution mouse lung imaging is challenging due to the signal loss caused by significant diffusion-weighting by the imaging gradients, particularly in larger airways where <sup>3</sup> He diffusion is maximized. In this chapter, a modified FGRE approach called X-Centric is described for acquiring <sup>3</sup> He mouse lung MRI. X-Centric is a center-out technique, allowing high-spatial-resolution, and high signal-to-noise ratio density-weighted imaging, as it is a short-TE method minimizing diffusion decay. Here, we also take advantage of a high-performance insertable-gradient-set interfaced with a clinical MRI system and a custom-built constant-volume ventilator to get the maximum benefits of X-Centric. High-spatial-resolution <sup>3</sup> He X-Centric imaging was performed in a phantom and mouse lungs, yielding a nominal resolution of 39 μm and 78 μm respectively. We also demonstrate the feasibility of 129Xe/19F X-Centric MRI in a phantom and in rat lungs.

Keywords: X-Centric, <sup>3</sup> He, 129X, 19F mouse, rat, lung, short-TE, x-nuclei, pre-clinical, hyperpolarized, MRI, COPD, CF, BPD, alpha-1 antitrypsin deficiency

#### 1. Introduction

Inhaled hyperpolarized gas lung MRI [1] was proven to be useful for the observation and treatment planning of several pulmonary diseases including chronic obstructive pulmonary disease (COPD) [2–6], asthma, [7–11] and lung cancer [12]. With the combined economic burden of COPD and asthma in Canada, Ontario being \$5.7 billion (2011) [13], and lung cancer being the leading cause of cancer deaths worldwide, [14] there has been a growing interest in developing new lung imaging techniques such as hyperpolarized <sup>3</sup> He and 129Xe MRI as well as hyperpolarized propane ( 1 H) MRI [15, 16] and thermally polarized inert fluorinated gas ( 19F) MRI [17–19] to better understand various lung diseases.

Our main focus will be on <sup>3</sup> He gas MRI of small animal models of lung disease [20, 21], as it is a promising tool to quantify airway and ventilation abnormalities associated with bronchoconstriction, airway narrowing, and subsequent bronchodilation [22, 23]. Small animal <sup>3</sup> He lung MRI normally assumes high spatial resolution imaging (<500 μm [22]) in order to obtain acceptable images for the quantitative analysis and visualization of rodent airways and lung parenchyma. However, high-resolution hyperpolarized gas MR imaging has a number of signalto-noise ratio (SNR) limitations due to the non-renewable nature of the nonequilibrium magnetization, requiring a minimized number of small flip angle radiofrequency (RF) pulses and quite short image acquisition times during a breath-hold which should be as short as a second in duration for the case of mouse lung imaging. Coupled with the issues of high-resolution imaging are the effects of fast diffusion of <sup>3</sup> He (˜<sup>2</sup> cm2 /s) through strong spatial encoding gradients, which can completely destroy the MRI signal in the airways and significantly attenuate the signal in the lung parenchyma [24–26]. There is a need for an imaging method which incorporates specific hardware and novel rapid image acquisition approaches that minimize the signal loss due to all of these mitigating factors and permit fast and high spatial resolution <sup>3</sup> He MR imaging of small animal airways and lung microstructure [27]. Such imaging modalities can be used for observation of ventilation heterogeneity, such as in an ovalbumin asthmatic mouse model [22, 27], as well as for pediatric lung applications [28] often requiring much smaller field-of-view (FOV).

The projection-reconstruction (PR) approach is the well-known [29] acquisition method that minimizes <sup>3</sup> He diffusion-induced signal decay, as it acquires the k-space center immediately following the excitation pulse and before the imaging gradient starts which avoids significant signal loss occurs [27]. Both PR and spiral/ cones acquisitions belong to free induction decay based or apparent transverse relaxation time (T2\*) based methods, acquiring k-space in a non-Cartesian fashion. It has been demonstrated that those methods reduce diffusion-weighting for <sup>3</sup> He MRI of rodent lungs [22, 30–34]. In addition, they ensure very short echo-times ((TE) < 200 μs) which minimizes T2\*-weighting, leading to pure gas density weighting [35]. Unfortunately, PR uses a significant number of RF pulses (50,640 views for a 128 ° 128 ° 128 matrix size and 204,196 for a 256 ° 256 ° 256 matrix size [36]) or radial projections to sufficiently sample k-space and this naturally leads to reduction of the available SNR from the non-renewable hyperpolarized state [27]. Cones [34] and spiral [28] k-space trajectories normally do not require a significant number of RF tipping or interleaves to fill k-space, but this should still be compensated by using an elongated readout-window which can be difficult to achieve due to the short T2\* of <sup>3</sup> He gas in small animal lungs (T2\* < 3 ms, at high field [37]). Commonly, non-Cartesian acquisitions require dedicated k-space interpolating and regridding as well as density-weighting procedures for image reconstruction [27], which often can have smoothing effects and resolution implications. Furthermore, non-Cartesian traversal of k-space by either PR or spiral/cones often requires significant oversampling, which leads to commensurate increases in acquisition time and consequently prohibiting single breath-hold imaging in rodents. Driehuys et al. [22] suggests that <sup>3</sup> He MRI measurements of rapid dynamic changes in mouse lungs followed by methacholine challenge can be done more time and cost efficiently with fast 2D Cartesian imaging compared to 3D PR.

An alternative to PR (which may suffer from poor edge resolution and sampling/ reconstruction artifacts [18]) for high spatial resolution single-breath 2D imaging of small animal lung employs two separate excitations with inverted read-out gradients to acquire both halves of Cartesian k-space in a center-out fashion to minimize both diffusion-weighting [38, 39] and T2\*-weighting [17]. This technique is known as the X-Centric method [27, 39]. It uses a well-known Fast Gradient-Recalled Echo (FGRE) sequence as its basis [27, 39], so the method can be relatively easy implemented on any clinical MRI scanners [17, 27]. Also, because X-Centric is a Cartesian method, it does not require interpolating, regridding and densityweighting procedures for image reconstruction [17, 27, 39]. It minimizes both

#### High Resolution <sup>3</sup> He Pulmonary MRI DOI: http://dx.doi.org/10.5772/intechopen.84756

gradient-induced diffusion attenuation [27, 39] and T2\*-based signal attenuation [17] at the center of k-space, and does not depend on the image resolution for a chosen FOV and bandwidth (BWread).

In this chapter we present an X-Centric approach developed for a single breathhold high spatial resolution <sup>3</sup> He MRI of mouse lungs. The method takes advantage of a high performance insertable gradient set interfaced with a clinical MRI system and precise custom built constant-volume ventilator. The X-Centric approach is compared to partial-echo FGRE [25] on the basis of SNR efficiency over a range of spatial resolutions in a phantom [27]. The robustness of the X-Centric technique for 3 He in-vivo lung imaging of mice is demonstrated [27]. The feasibility of the X-Centric approach to image dissolved hyperpolarized 129Xe in a phantom and with 19F gas in a rat lung [17] is also demonstrated in this chapter.

#### 2. Theoretical background

#### 2.1 <sup>3</sup> He MRI

Table 1 summarizes the physical properties of the <sup>3</sup> He isotope along with other gases used for inhaled lung MRI [40] such as 129Xe [1, 41], inert fluorinated gases possessing 19F spins [17–19] and propane [16] which is proton-based ( 1 H). It is wellknown, that helium gas has the highest self-diffusion coefficient (D), which is around 2 cm<sup>2</sup> /s for pure helium, and it is around 0.9 cm<sup>2</sup> /s, when helium diffuses in air or is part of a helium/air mixture. Its very high diffusivity suggests that highresolution <sup>3</sup> He MRI can be quite challenging due to the induced decay of the MR signal by the applied imaging gradients [22, 24]. However, signal attenuation due to diffusion is not the only reason making high-resolution <sup>3</sup> He MRI problematic. In most cases it requires very careful consideration of all MRI related physical properties of <sup>3</sup> He and sequence parameter effects, such as the contribution of relaxation times to both signal decay and image blurring [42]. In addition to these


SF6 and <sup>3</sup> He T2\* values shown within major airway<sup>a</sup> and parenchema<sup>b</sup> of the rodent lung. 129Xe T2\* values shown for xenon gasa within the major airway and xenon dissolved<sup>c</sup> within lung tissue (barrier) and red blood cells. 129Xe and <sup>3</sup> He T<sup>1</sup> values<sup>d</sup> measured in rodent lungs after several wash-out (anoxic) breaths of the hyperpolarized gas. Cost of the 86% enriched 129Xee .

#### Table 1.

Comparison of <sup>3</sup> He, 129Xe, inert fluorinated gas and propane. considerations, one needs to consider the use of special MRI hardware such as a high-performance high-slew-rate imaging gradient set [43] and a very high precision ventilator [27] for in-vivo high-resolution imaging. The high diffusivity of <sup>3</sup> He as well as its short relaxation times and MRI hardware limitations should all be taken into account for the proper choice of the pulse sequence when high spatial resolution <sup>3</sup> He MRI is the goal.

#### 2.2 Relaxation times

Generally, both longitudinal and apparent transverse relaxation [44] times (T<sup>1</sup> and T2\*) of <sup>3</sup> He are not physical constants and they both depend on the main magnetic field strength, its gradients, presence of paramagnetic impurities interacting with the hyperpolarized gas (e.g., oxygen), and finally, diffusivity. Normally, the main mechanism for T<sup>1</sup> decay of hyperpolarized gases (i.e., polarization losses in contrast to conventional thermally polarized spins) is the presence of oxygen in lung [45]. The T<sup>1</sup> decay rate is inversely proportional to the partial oxygen pressure (pO2) [46]:

$$\frac{1}{T\_1} = \frac{1}{T\_{1,0}} + \kappa \int\limits\_{0}^{time} pO\_2(t)dt\tag{1}$$

where T1,0 is the longitudinal relaxation time expected in the absence of O2 and κ is the relaxivity of <sup>3</sup> He due to the presence of O2 (0.38 atm�<sup>1</sup> s �<sup>1</sup> at body temperature [46]).

This suggests that a small number of anoxic pre-breaths or wash-out breaths prior the actual <sup>3</sup> He MRI measurements can be helpful for minimizing the oxygen concentration in lung prior to acquisition and administration of hyperpolarized gas. Table 1 suggests that the T<sup>1</sup> in small animal lungs after applying a number of wash-out breaths can be sufficiently long [47] compared to the typical image acquisition times of 1 s [27] for mice and 10 s for rats [37]. As previous work shows, eight wash-out breaths drastically reduce the O2 concentration [17, 23, 48]. The value of T2\* for <sup>3</sup> He normally varies between the major airways (1.5 � 0.25 ms [37]) and lung parenchyma (3.0 � 1.0 ms [37]) due to the differences in airway size which have a significant effect on the mode of diffusion (e.g., free diffusion vs. restricted) and amount of signal loss due to diffusion. Table 1 shows T2\* values obtained for hyperpolarized gas within rat lungs at 3 Tesla. As expected T2\* estimates obtained for the major airways were smaller than the estimates observed in lung parenchyma.

#### 2.3 Signal to noise ratio

An imaging sequence should be optimized to simultaneously ensure maximum SNR and minimal blurring in order to successfully perform high-resolution hyperpolarized <sup>3</sup> He MR measurements. Understanding of the mechanisms that affect SNR and contribute to image blurring is the basis for the developing an imaging technique appropriate for hyperpolarized gas imaging. Generally, an expression for the SNR of a FGRE sequence can be expressed as [42]:

$$\text{SNR} \propto \Delta x \Delta y \Delta x \sqrt{\frac{N\_x N\_y N\_z}{BW\_{rad}}} f\left(\rho, \text{ } P, a, \text{ } T\_2^\*, \text{ } T\_1, \text{ } b, \text{ } D\right) \tag{2}$$

#### High Resolution <sup>3</sup> He Pulmonary MRI DOI: http://dx.doi.org/10.5772/intechopen.84756

� � where ΔxΔyΔz is the voxel size, NxNyNz is the number of k-space samples in each direction, BWread readout bandwidth and f ρ; P; α; T<sup>2</sup> <sup>∗</sup>; T1; b; D is the pulse-sequence-dependent function that determines the signal amplitudes at readout for the center of k-space [49]:

$$f\left(\rho,\ P,\ T\_1,\ T\_2^\*\ a,\ b,\ D\right) \approx \rho \mathbf{P} \sin\left(a\right) \exp^{\left(-T\mathcal{E}/T\_2^\*\right)} \exp^{\left(-T\mathcal{R}/T\_1\right)} \exp^{\left(-b\ \ D\right)}\tag{3}$$

As it can be seen, the signal amplitudes at readout depend on the density (ρ) and polarization (P) of <sup>3</sup> He, flip angle (α), TE, repetition time (TR) and finally, diffusion attenuation [50] introduced by slice select and/or frequency encoding gradients. In order to simultaneously minimize T<sup>1</sup> decay and T2\*-induced blurring and maximize signal, one should use a short-TE (TE < 1 ms) center-out k-space traversal pulse sequence to ensure only density-weighting at the center of k-space. For the case of TE < T2\*, and if the total acquisition time (1–10 s) is much less than T<sup>1</sup> (25–90 s) the pulse sequence dependent function can be expressed as:

$$f(\rho, \ P, \ a, \ b, \ D) \approx \rho P \sin \left( a \right) \exp^{\left(-b \ \ D \right)} \tag{4}$$

where b is the b-value, the diffusion weighting factor that depends on the diffusion gradient duration and magnitude (which in this case are imaging gradients); and D is the diffusion coefficient of <sup>3</sup> He in the lung (0.1–2 cm<sup>2</sup> /s) [37, 50].

#### 2.4 Flip angle

Let us consider each factor in Eq. (4) individually in order to optimize this function. As the hyperpolarized magnetization is not renewable due to the its nonequilibrium nature, it needs to be spent (flipped by RF pulses) very efficiently. A single 90<sup>o</sup> RF pulse can effectively waste all of the available magnetization. We start with the two approaches of setting flip angles in hyperpolarized gas MRI. The first method is the Constant Flip Angle (CFA) approach [51] or small flip angle approach (1<sup>o</sup> < flip angle <10<sup>o</sup> ). Using a CFA approach, the optimal flip angle (αopt), which provides the highest signal for sequential phase encode ordering, is expressed as [51]:

$$a\_{opt} = \begin{array}{c} \tan^{-1} \sqrt{2/\left(N\_{\mathcal{V}} N\_{x}\right)} \end{array} \tag{5}$$

Unfortunately, the use of the optimal flip angle given by Eq. (5) leads to significant signal decay due to the constant-value RF pulses applied during k-space acquisition. As an example, for a 2D case where Ny = 128 and αopt = 7.1<sup>o</sup> the signal decays by �60% across the k-space). An undesirable consequence of such decay is reduced image resolution as result of the RF pulse "history" during k-space acquisition. Point-spread-function (PSF) simulations for a 2D case with center-out phase encode ordering (Figure 1) confirmed that the CFA approach leads to image blurring (Full Width at Half Maximum (FWHM) = 1.5 pixels) [47]. This signal decay-induced image blurring can be eliminated by using a Variable Flip Angle (VFA) approach (1<sup>o</sup> < flip angle <90<sup>o</sup> ) [52]. VFA mitigates signal loss during image acquisition (i.e., blurring) by ensuring a constant signal for each phaseencode line by starting with a low flip angle and incrementing the flip angle of each RF pulse, line after line [52]. The flip angle for each i th phase-encode line of a VFA sequence can be calculated from the following equation [47]:

$$a\_i = \tan^{-1} \left[ \frac{\exp \left( - (N\_\mathcal{V} N\_x - i) T R / T\_1 \right)}{\sqrt{N\_\mathcal{V} N\_x - i}} \right] \tag{6}$$

For the case where the total image acquisition time is much less than T<sup>1</sup> (T<sup>1</sup> typically >10 sec under normal conditions), Eq. (6) takes the simpler form [47]:

$$a\_i \approx \tan^{-1} \left[ \frac{1}{\sqrt{N\_\mathcal{V} N\_x - i}} \right] \tag{7}$$

For a 2D case (Ny = 128), α<sup>1</sup> = 5.1o and α<sup>128</sup> = 90<sup>o</sup> , producing no CFA-like signal decay, and consequently, no image blurring due to RF pulse "history" [47]. Simulations have shown that VFA only shows blurring of 1.2 pixels due to discrete sampling [47] and because of these benefits this method is preferable for high spatial resolution <sup>3</sup> He imaging.

#### 2.5 b-Value

Eq. (4) suggests that signal decay due to diffusion is the main reason the MR signal degradation (assuming minimal T<sup>1</sup> and T2\* induced blurring and signal loss) and consequently poses as the main difficulty for high spatial resolution <sup>3</sup> He imaging. Figure 2 shows the calculated NMR signal as a function of b-value for two different diffusion coefficients, typical for a small animal airways and lung parenchyma respectively [37]. It can be seen that for a b-value >3.5 s/cm<sup>2</sup> , the signal intensity in the airway is almost zero and there is a 60% loss of signal for lung parenchyma. Let us consider the dependence of b-value on the diffusion gradient duration and magnitude. In general, the b-value for any gradient waveforms can be calculated from the first principles [42]:

#### Figure 1.

Point spread function simulation for the CFA approach (a) the k-space is simulated for hyperpolarized CFA <sup>3</sup> He signal of a 128 � 128 pixel object with αopt = 7.12° (Eq. (5)) and infinite T1. (b) Fourier transform of (a) revealing the PSF [47]. The calculated FWHM of the PSF indicates significant blurring (1.5 pixels) in the phase-encode direction [47]. The observed blurring is due to both RF pulse history of non-recoverable magnetization and discrete sampling [47]. Variable flip angle only shows blurring of 1.2 pixels due to discrete sampling [47]. Figure adapted from Ouriadov et al. [47].

High Resolution <sup>3</sup> He Pulmonary MRI DOI: http://dx.doi.org/10.5772/intechopen.84756

#### Figure 2.

Dependence of NMR signal on b-value for two different diffusion coefficients of <sup>3</sup> He calculated from the diffusion term of Eq. (4). The dashed line curve represents signal obtained for a D value that is typical for airways (free diffusion) and the solid line curve represents signal obtained for a typical D value in lung parenchyma. Plot shows that for a b-value >3.5 s/cm<sup>2</sup> , the signal intensity in the airway is almost zero and there is 60% of signal loss for lung parenchyma.

where γ is gyromagnetic ratio, G is gradient amplitude and t, t<sup>1</sup> is gradient duration. Unfortunately, Eq. (8) cannot be analytically simplified further for a gradient waveform specific to the frequency encoding imaging gradient. For a relatively simple diffusion gradient waveform such as two bipolar rectangular pulses, the b-value can be calculated based on the following equation [53]:

$$b = -\frac{2}{3}\gamma^2 G^2 \delta^2 t\_d \text{ with } t\_d = \left[\Delta - \frac{\delta}{3}\right],\tag{9}$$

where δ is the diffusion gradient duration, Δ is a distance between two gradient pulses and t<sup>d</sup> is the diffusion observation time. Note, that b-value strongly depends on the gradient duration (δ<sup>2</sup> t<sup>d</sup> = time<sup>3</sup> ) then the magnitude (G<sup>2</sup> ), which suggests that minimizing the gradient duration reduces the b-value and consequently minimizes the signal loss. This is especially critical for high-resolution <sup>3</sup> He imaging, as the imaging gradient magnitude cannot be really decreased when trying to achieve high spatial resolution, or small FOV for mice.

#### 2.6 X-centric

As <sup>3</sup> He gas has the highest self-diffusion coefficient of the usable hyperpolarized gases and it is a physical constant, the diffusion weighting b-value due to the application of frequency encoding gradient or x-gradient should be minimized as

much as possible. A full-echo FGRE pulse sequence ensures the highest b-value (and subsequent signal loss) for gradient echo based sequences [22, 25]. The use of a partial-echo (partial in the x-gradient direction, 62.5% sampled points) FGRE pulse sequence can significantly reduce the b-value compared to the full echo FGRE due to reduction in readout time; however, even this reduced b-value can be still large enough (>2.0 s/cm<sup>2</sup> ) to completely destroy MR signal in some lung geometries such as in the case of high spatial resolution <sup>3</sup> He imaging in small animals [22, 27]. The main limitation of FGRE is the undesirable dependence of the b-value on the xgradient making this sequence not very suitable for high-resolution <sup>3</sup> He imaging. However, it is possible to modify the original FGRE sequence, so it ensures minimal b-values due to x-gradients and full independence of the b-value from the image resolution for the chosen field-of-view (FOV) and BWread. A modified or X-Centric FGRE pulse sequence is presented in Figure 3. The idea of the X-Centric FGRE pulse sequence is straightforward. It acquires half of an echo in the x-direction by starting at the center of k-space then repeats the same k-space line but with the xgradient inverted in order to collect the second half of the echo in x-direction. Figure 4 shows part of an experimental x-gradient waveform (nominal resolution 78 μm) from the start of the line acquisition until the center of the k<sup>x</sup> line for partialecho (open squares and dashed line) and X-Centric (solid squares and solid line) FGRE pulse sequences. The area of the pre-phasing gradient shown on the plot is equal to area of the main readout gradient lobe for both acquisition methods. Figure 4 suggests that the X-Centric FGRE should decrease b-value at k<sup>x</sup> = 0 due to the much shorter diffusion time and smaller pre-phasing gradient magnitude leading to higher signal at the center of k-space.

The X-Centric approach requires two discrete acquisitions for each k-space line making the method twice as long in acquisition time compared to a conventional partial-echo FGRE approach for a chosen spatial resolution and matrix size [17, 27, 39]. Though, this is not a significant cost to achieve high-resolution imaging of the lungs of small animals, the scan time can be further reduced by using a partial-echo method in the phase-encoding direction [17, 27, 39]. Note, that the use of a partial-echo method in the phase-encoding direction makes the X-Centric scan time only 10% longer than the partial-echo FGRE sequence for a similar spatial resolution and matrix size [27].

#### Figure 3.

Pulse sequence timing diagram of X-centric acquisition scheme. The X-centric approach minimizes signal loss due to diffusion since the number of k-space points acquired before the center of the kx line is significantly reduced. Figure adapted from Ouriadov et al. [27].

#### Figure 4.

Expanded view of the readout gradient waveform in partial-echo FGRE (open squares and dashed line) and X-centric (solid squares and solid line) for acquisitions with <sup>256</sup> � <sup>256</sup> image matrix FOV <sup>=</sup> <sup>2</sup> � <sup>2</sup> cm<sup>2</sup> , and BWread = 62.5 kHz [27]. X-centric waveform has a shorter TE (mainly determined by the duration of the phase-encoding gradient) and much smaller area compared to the partial-echo FGRE waveform resulting in a decreased b-value at kx = 0 [27]. Both X-centric and FGRE gradient waveforms show the readout gradient until k<sup>x</sup> = 0 point only [27]. Figure adapted from Ouriadov et al. [27].

#### 2.7 Diffusion attenuation

The exact b-value for gradient waveforms presented in Figure 4 can be calculated from Eq. (8). Thus, b-values calculated for the partial-echo and X-Centric FGRE (256 � <sup>256</sup> matrix size) were 2.29 and 0.061 s/cm<sup>2</sup> respectively. Figure <sup>2</sup> indicates that a b-value more than 2.0 s/cm<sup>2</sup> leads to the complete signal decay in the major airway which is a significant limitation for high spatial resolution <sup>3</sup> He imaging of the mouse lung. In contrast, the X-Centric sequence should theoretically ensure minimal signal decay for given diffusion coefficients. Note that for the unrestricted <sup>3</sup> He diffusion case the theoretical SNR (SNRtheor) should depend only on b-value and D:

$$\text{SNR}\_{theory} = \text{S}\_0 \exp^{(-bD)} \tag{10}$$

where S<sup>0</sup> is the initial signal (S<sup>0</sup> = 1), D is the free <sup>3</sup> He self-diffusion coefficient (2.0 cm2 /s) and b takes on a value determined by the chosen FOV and BWread.

According to Eq. (2), for the case of the high-resolution 2D <sup>3</sup> He phantom imaging, the SNR should generally depend on the diffusion attention, image resolution, BWread and the first flip angle in the VFA scheme [27]:

$$\text{SNR} \propto P\rho \sin \left(\alpha\_1\right) \cdot \Delta x \Delta y \sqrt{\frac{N\_\text{x} N\_\text{y}}{BW\_{read}}} \exp^{\left(-bD\right)}\tag{11}$$

as Δx, Δy can be expressed through FOVs and Nx, and Ny Eq. (11) can be rewritten as

$$\text{SNR} \propto P\rho \sin \left( a\_1 \right) \quad \frac{\text{FOV}\_x \text{FOV}\_y}{N\_x N\_y} \sqrt{\frac{N\_x N\_y}{BW\_{read}}} \exp^{\left(-bD\right)} \tag{12}$$

If one keeps FOV, BWread, <sup>3</sup> He polarization and volume constant across all phantom measurements, then the SNR comparison between the partial-echo and X-Centric FGRE can be done by normalizing the experimental SNR by the first pffiffiffiffiffiffiffiffiffiffiffiffi flip angle in the VFA scheme (α1) and matrix size ( NxNy). Thus, for a SNR normalized by the respective first applied flip angle and matrix size, one can write the following expression [27]:

$$\text{SNR}\_{nor} = \frac{\text{SNR}}{\sin \left(a\_1\right)} \sqrt{N\_\text{x} N\_\text{y}} = \text{Const} \cdot \exp^{\left(-bD\right)}\tag{13}$$

where Const ¼ Pρ <sup>p</sup> FOVffiffiffiffiffiffiffiffiffiffiffi xFOVy . This final SNRnor equation can be used for SNR BWread comparisons of the phantom images obtained for different resolutions with the partial-echo and X-Centric FGRE methods; also this equation is similar to Eq. (10).

#### 2.8 Phantom imaging

A plastic 10 mL syringe filled with hyperpolarized <sup>3</sup> He was used to validate the X-Centric sequence for high spatial resolution imaging [27]. Figure 5 shows a 2D coronal view whole projection phantom images obtained for partial-echo (top panel) and X-Centric (bottom panel) FGRE for five different resolutions. Images start from 64 � 64 matrix size and end with 512 � 512 from left to the right [27]. The images indicate that at low nominal resolution (625 μm, 64 � 64 matrix size) there is no qualitative or visual difference in the SNR between the images obtained for the partial-echo and X-Centric FGRE. However, there is virtually no signal in phantom images obtained with partial-echo FGRE when nominal resolution is

#### Figure 5.

Hyperpolarized <sup>3</sup> He phantom images (64 � 64, 128 � 128, 256 � 256, 384 � 384, 512 � 512 from left to right with <sup>a</sup> FOV <sup>=</sup> <sup>2</sup> � <sup>2</sup> cm<sup>2</sup> , and BWread = 62.5 kHz) obtained in the coronal plane at 1 atm gas pressure (D = 2.0 cm2 /s) for image acquisition times up to 6.7 s for: (a) partial-echo FGRE (top row with b-values ranging from 0.1 to 7.04 s/cm<sup>2</sup> at kx = 0) and (b) X-centric (bottom row with a b-value equal to 0.061 s/cm2 at kx = 0) [27]. Whereas the X-centric images demonstrate only a gradual loss in signal intensity with increasing image matrix, the partial-echo FGRE had no signal intensity for images with matrix sizes of 384 � 384 and 512 � 512. The partial-echo FGRE image obtained for 256 � 256 has been filtered with a Hanning filter to increase signal intensity. The changes of syringe vertical position are due to the slight displacement of the syringe inside the holder between fresh gas administrations. Figure adapted from Ouriadov et al. [27].

#### High Resolution <sup>3</sup> He Pulmonary MRI DOI: http://dx.doi.org/10.5772/intechopen.84756

greater than 256 ˜ 256 (78 μm nominal resolution). As expected, the signal intensity of the images obtained with X-Centric gradually decays with increasing nominal resolution and decreasing flip angle (because the first pulse for VFA is a function of resolution, Eq. (7)) but even for 512 ˜ 512 matrix size (39 μm nominal resolution) it does not go below the noise floor as is seen in the case of the partial-echo FGRE. The SNRnor dependence on resolution (bottom axis) and/or b-value (top axis, partialecho case only) for the phantom images obtained with the partial-echo (solid squares and solid line) and X-Centric FGRE (solid circles and solid line) is plotted in Figure 6. Clearly, the SNRnor obtained for the X-Centric did not depend on image resolution, i.e., diffusion attenuation was minimal even for very high image resolution (39 μm). The experimental phantom results were consistent with the predicted SNR (SNRtheor) calculated based on Eq. (10) for the partial-echo (open squares and dashed line) and X-Centric FGRE (open circles and dashed line). Eqs. (10) and (13) indicate that for the case of unrestricted <sup>3</sup> He diffusion (D = 2 cm2 /s) one should observe minor signal loss (°10%) using X-Centric and almost 100% signal loss using the partial-echo FGRE for a nominal resolution greater than 78 μm. Thus our consideration of the factors determining the SNR for high spatial resolution <sup>3</sup> He MR measurements shows that in the case of relatively short TR and TE (pulse sequence parameters for 256 ˜ 256 matrix size are present in Table 2) diffusion is the major reason for signal loss. Using the X-Centric method, diffusion decay due to the x-gradients is significantly minimized. It should be noted that we observed a slight

#### Figure 6.

Scaled normalized SNR of the phantom images (Eq. (13)) shown in Figure 5a and b versus spatial resolution (bottom horizontal axes) and b-value (top horizontal axes) for: (i) partial-echo FGRE (solid circles and solid line) and X-centric (solid squares and solid line) acquisitions [27]. These data are consistent with the predicted data (open squares and dashed line for partial-echo and open circles and dashed line for X-centric) based on the calculated b-values (Eq. (13)) [27]. This plot shows independence of normalized SNR from image resolution for the x-centric acquisition schemes [27]. The b-value axis (calculated for experimental x-gradient waveform based on Eq. (8)) applies only to the partial-echo FGRE curves as the b-values for X-centric curves were constant (0.061 s/cm<sup>2</sup> ) as a function of resolution [27]. Figure adapted from Ouriadov et al. [27].


VFA = variable flip angle; CFA = constant flipped angle; EA = Ernst angle; FOV = field of view; TR = relaxation time; TE = echo-time.

#### Table 2.

MRI acquisition parameters for methods in this chapter.

decrease of the SNRnor (around 10%) for a nominal resolutions better than 75 μm (Figure 6) which were likely due to T2\* decay because as resolution increases, TE becomes longer due to the longer duration of phase-encoding gradient. Nevertheless, X-Centric phantom images show no evidence of image blurring due to the RF pulse history or T2\*-based signal decay in the frequency/phase encoding directions. The phantom measurements have confirmed that significant diffusion attenuation induced by the x-gradient with the partial-echo FGRE sequence can be the main restriction for high spatial resolution <sup>3</sup> He imaging of the lung.

#### 3. High spatial resolution imaging of mouse lung

#### 3.1 Animal preparation

The animal protocol was approved by the Animal Use Subcommittee of the University Council on Animal Care at the University of Western Ontario, London, Ontario, Canada and the Animal Care Committee at Merck Frosst Canada, Kirkland, Quebec, Canada. Each mouse (BALB/c) was pretreated with Midazolam (1 mg/kg) intraperitoneally (i.p.) 5 min before anesthesia with ketamine (95 mg/kg, i.p.) and xylazine (6.4 mg/kg, i.p.). Following anesthesia, the tail vein was cannulated (26 GA Abbott catheter) to maintain anesthesia with ketamine (30 mg/kg) and xylazine (2 mg/kg) every 40 min. To allow artificial mechanical ventilation, the trachea was cannulated by tracheotomy using an 18 GA Teflon catheter. A line for administration of pancuronium (0.8 mg/kg, i.p.) was established using a 26 GA cannula to allow for injections after the mouse was being ventilated inside the scanner. Other physiological instrumentation included ECG and a rectal temperature probe (SA Instruments Inc., Stony Brook, NY).

#### 3.2 Ventilation

It was a very challenging task to design a ventilator to ventilate mice inside an MRI because of the restrictions due to the high magnetic field environment and the requirement to accurately deliver very small tidal volumes of gas to the lungs of mice at a high ventilation rate. Under spontaneous breathing, the mouse maintains its oxygenation with a tidal volume less than 0.1 ml [54, 55]. Because of the dead

#### High Resolution <sup>3</sup> He Pulmonary MRI DOI: http://dx.doi.org/10.5772/intechopen.84756

space volume within the tubing necessary for connecting mouse airways to the ventilator, the mouse was normally ventilated with tidal volume of 0.2 ml (8 ml/kg). In this range, the amount of the ventilating gas (or fresh gas) getting compressed in the gas line could significantly reduce the portion of fresh gas entering the lungs for gas exchange. Without a proper correction for this effect, adequate ventilation would not be achieved at this ventilating setting. In many cases, the ventilation was set with a much larger tidal volume (>10 ml/kg) at a slower ventilation rate (60–90 bpm). Since lung volume under spontaneous breathing is only 0.3–0.4 ml [56], the larger tidal volume approach could overextend the airways. As a result, many mice were observed to expire shortly after the ventilation, impeding the completion of studies.

To resolve these problems, we modified a flexiVent ventilator (Figure 7, Scireq, Montreal, QC, Canada) that incorporated real-time monitoring of volume and pressure to allow corrections of the gas compression effect. To minimize the gas line volume and ensure the effectiveness of ventilation, a non-metal swing valve was designed and placed close to the animal (inside the scanner) to regulate the gas flow. Different than the pneumatic valves normally used in the magnetic environment, the swing valve was connected to two driving solenoid valves (which reside far away from the high magnetic field) via two multiple-section light-weight carbon fiber rods. This mechanical design enabled fast open/close activations. The activation of a solenoid valve pulls the swing valve to close the ports on one side and open the ports on the other side. Two solenoid valves were activated in sequence to generate a swing motion that opens and closes the ports on the respective sides alternately. This swing motion was synchronized with the plunging motion of the piston that drives the gas into the lung to form a ventilation cycle. Because of the fast response of the swing valve, the modified flexiVent ventilator is capable of performing the forced oscillation method to access pulmonary mechanics. To minimize oxygen-induced depolarization (T<sup>1</sup> decay) of hyperpolarized <sup>3</sup> He gas, the flexiVent ventilator was modified to deliver breathing gas and hyperpolarized <sup>3</sup> He gas separately using two individual syringes. The ventilation system was integrated to the MR scanner (Figure 7) using custom-designed LabView software (National Instruments) to allow ventilation-synchronous acquisition. Following the animal preparation, the animal was connected to the ventilator and ventilated using a tidal volume of 8 ml/kg at 100 bpm and was placed inside the scanner. The breath-hold time for imaging was 1 s.

Figure 7. GE 3 T clinical MRI scanner using an external gradient set for rodents and the integrated custom-built flexiVent ventilator for mice.

#### 3.3 <sup>3</sup> He polarization and delivery

3 He gas was polarized using an optical pumping spin-exchange system (Helispin™, GE Healthcare, Durham, NC). The <sup>3</sup> He gas was typically polarized for at least 8 hours to achieve a polarization over 35%. Prior to transferring the hyperpolarized <sup>3</sup> He gas to a 150 mL Tedlar bag (Jensen Inert Products, Coral Springs, FL), the bag was carefully washed with N2 gas three times to eliminate any oxygen contamination. Following the gas filling, the <sup>3</sup> He bag was placed inside a custom-made plexiglass chamber and connected to a <sup>3</sup> He intake line of the ventilator. The chamber was then placed within the homogenous B0 field inside the scanner to conserve <sup>3</sup> He polarization (T<sup>1</sup> = ˜43 min). To facilitate gas intake by the ventilator, the pressure in the chamber was kept at 3 cm-H2O during the gas delivery.

To validate the X-Centric imaging method, 2 mL at 1 atm of the hyperpolarized 3 He gas was drawn into a 10 mL syringe for phantom testing [27].

#### 3.4 MRI hardware and sequence parameters

Hyperpolarized <sup>3</sup> He MR imaging was performed on a GE clinical scanner (3 T, Excite 12.0) which was converted using a home-made insert gradient set (Figure 8, maximum gradient at 50 G/cm, 17 cm in diameter, slew rate = 2100 mT/m/s) [43] to allow high spatial resolution imaging for mice. This high performance insert gradient set was used for phantom and mouse imaging [27]. A quadrature birdcage RF coil for mice (Figure 9, 3 cm in diameter and 6 cm in length, Morris Instruments, Ottawa, ON) was used for <sup>3</sup> He imaging (97.3 MHz) [27]. The coil was driven using a 8 kW AMT 3 T90 RF power amplifier (GE Healthcare, Waukesha, WI)

#### Figure 8.

High performance water cooled insertable gradient set with maximum gradient at 50 G/cm, 17 cm in diameter and slew rate = 2100 mT/m/s used for high-resolution <sup>3</sup> He imaging.

#### Figure 9.

A quadrature birdcage RF coil for mice (Morris instruments, Ottawa, ON): 97.3 MHz, 3 cm in diameter and 6 cm in length, was used for used for high-resolution <sup>3</sup> He imaging.

using a manufacturer-supplied T/R switch and preamplifier tuned to 97.3 MHz. Typical rectangular 90° RF pulse lengths were 100 μs [27]. A VFA RF pulse trajectory was employed to reduce blurring due to RF de-polarization [47, 52]. Center-out sampling in the phase-encoding direction was used [52]. The VFA pulses were calibrated as described in [48].

#### 3.5 Image reconstruction

A custom-made IDL6.4 (ITT Visual Information Solutions, Boulder, CO) routine was used for off-line Fast Fourier transformation (FFT) of the k-space data. For reconstruction of the X-Centric data, a combination of both halves of k-space was applied prior to the inverse Fourier transformation as follows. To form a full line of k-space data in the read-out direction, the two half-echo data sets were combined [17]. Each half-echo data set included two additional points (BWread = 62.5 kHz dwell time was 8 μs) prior to k<sup>x</sup> = 0 which can be thought of as partial-echo factor of 0.505 for each half.

Acquisition of a few extra k<sup>x</sup> data points is helpful since the presence of background gradients can shift the origin of k-space by one or more points resulting in mis-sampling or causing image artifacts [57]. These extra k-space points (only two) were removed before combining the two halves of k-space prior to reconstruction [17]. The disadvantage of this approach is that this leads to a b-value three times larger and longer TR due to the extra sampled points. However, phantom results demonstrate that this increased b-value (b = 0.061 s/cm2 ) is still quite practical, ensuring minimal mis-sampling even in the presence of significant backgroundgradients and gradient-induced eddy currents [17, 27, 39]. Collection of the extra four k<sup>x</sup> points (two points per half) for each k<sup>x</sup> line results in a 1.5% over-sampling factor for a 256 ˜ 256 matrix size as an example [27].

#### 3.6 In-vivo imaging

X-Centric sequence parameters for the highest image resolution (78 μm) are shown in Table 2. In-vivo high-resolution 2D whole lung projection <sup>3</sup> He mouse lung images obtained for partial-echo (top panels) and X-Centric (bottom panels) FGRE acquisitions for three different matrix sizes are shown in Figure 10 (axial view) and Figure 11 (coronal view). Images were obtained during a 1 s breath-hold following three wash-out <sup>3</sup> He pre-breaths. It is easy to see that both the axial and coronal images of the mouse lung obtained at high spatial resolution (78 μm) with X-Centric demonstrates visible signal in the major airways. Axial and coronal images of mouse lung obtained for the same resolution setting with partial-echo FGRE show no signal in the trachea as well as lower SNR in the lung periphery. The mean SNR for X-Centric images computed from the mouse lung airways and parenchyma [58] were 25 and 14 respectively [27]. The corresponding mean SNR for the partial-echo FGRE computed for airways and parenchyma were 1.8 (which is less than the Rose criteria of SNR = 5, [59]) and 8 respectively [27]. The large b-value of the partial-echo FGRE led to complete signal destruction in the major airways where <sup>3</sup> He experiences nearfree diffusion (D = 2 cm2 /s). Besides this, a relatively long TE (1.2 ms [27]) in combination with a shorter T2\* (Table 1) in the mouse lung trachea also significantly contributes to signal decay together with diffusion losses. In comparison to the larger airways, the lung parenchyma within the images obtained with partialecho FGRE still shows some signal, which is not surprising considering that the <sup>3</sup> He diffusion was significantly restricted here (D = 0.21 cm2 /s [37]) and T2\* value was long enough (Table 1).

#### Figure 10.

Hyperpolarized <sup>3</sup> He mouse images (64 � 64, <sup>128</sup> � 128, <sup>256</sup> � <sup>256</sup> from left to right FOV <sup>=</sup> <sup>2</sup> � <sup>2</sup> cm<sup>2</sup> , and BWread = 62.5 kHz) obtained in the axial plane using a single breath hold with image acquisition times up to 1.1 s for: partial-echo FGRE (top panel, b-values ranging from 0.1 to 2.29 s/cm2 at kx = 0) and X-centric (bottom panel, b-value equals 0.061 s/cm2 at kx <sup>=</sup> 0) [27]. At high-resolution (256 � 256), the partial-echo FGRE images show no signal intensity in the trachea in contrast with the high-resolution X-centric images due to differences in the incidental diffusion weighting in the large airway compared to the lung parenchyma (shown by the arrow) [27]. Figure adapted from Ouriadov et al. [27].

The larger SNR in the major airway (X-Centric mouse lung images) compared to the lung parenchyma is likely due to the higher <sup>3</sup> He density in trachea and reasonably short TE used in X-Centric sequence (0.7 ms [27]).

It is useful to calculate the SNR efficiency (ϒ) of the X-Centric and partial-echo FGRE acquisitions was using the following equation [60]:

$$\Upsilon = \frac{\text{SNR} / (\Delta x \Delta y)}{\sqrt{T\_{scan}}} \tag{14}$$

where Tscan is the total imaging time (in s) and SNR was measured based on the mean of Region of Interests (ROI) placed in the rodent lung airways and parenchyma divided by the standard deviation measured in the background (noise) [58]. Thus, X-Centric was 13.4 times more efficient than the partial-echo FGRE sequence for the case of the major airways [27] and 1.7 times more efficient in case of MR imaging of parenchyma [27] (Figure 11).

#### 3.7 Limitations

One of the limitations of the X-Centric method is the need to use the FFT for image reconstruction from two half-echo k-space samples rather than a single k-space. Instead, one can use a single half-echo k-space and Partial Fourier Reconstruction (PFR) [61]. To test this idea we used the Projection Onto Convex Sets (POCS) method [62] to reconstruct a phantom image for 130 � 256 (2 + 128 � 256) matrix size (Figure 5b). Figure 12 shows phantom images obtained with PFR (left column) and FFT (right column) for three different matrix sizes. The following SNR values were obtained for the presented images: SNR = 39 (160 � 256, PFR); SNR = 54 (160 � 256, zero-filling in x-direction and then FFT); SNR = 45

#### Figure 11.

Hyperpolarized <sup>3</sup> He mouse images (64 ˜ 64, <sup>128</sup> ˜ 128, <sup>256</sup> ˜ <sup>256</sup> from left to right FOV <sup>=</sup> <sup>2</sup> ˜ <sup>2</sup> cm<sup>2</sup> , and BWread = 62.5 kHz) obtained in the coronal plane within a single breath hold with image acquisition times up to 1.1 s for: partial-echo FGRE (top panel, b-values ranging from 0.1 to 2.29 s/cm<sup>2</sup> at kx = 0) and X-centric (bottom panel, b-value equals 0.061 s/cm2 at kx <sup>=</sup> 0) [27]. At high-resolution (256 ˜ 256), the partial-echo FGRE images show no signal intensity in the trachea in contrast with the high-resolution X-centric images due to differences in the incidental diffusion weighting in the large airway compared to the lung parenchyma (shown by the arrow) [27]. Figure adapted from Ouriadov et al. [27].

(130 ˜ 256, PFR); and SNR = 58 (256 ˜ 160, zero-filling in y-direction and then FFT). The results confirm that the FFT with zero-filling approach gives better SNR than the PFR method for 160 ˜ 256 matrix (39 vs. 54). We have calculated the efficiency of half-echo acquisition (130 ˜ 256, SNR = 45) reconstructed with PRF and X-Centric (256 ˜ 160, 62.5% under-sampling in y-direction, SNR = 58) reconstructed with FFT in order to compare imaging methods. Thus, the calculated efficiency ratio is 0.87 (Eq. (14), with Tscan = 256 and 160 + 160 y-gradient steps for the half-echo and the X-Centric respectively). This shows that a half-echo acquisition is 13% less efficient than X-Centric, which reflects smaller SNR due to the 50.1% sampling vs. 62.5% sampling for the case of X-Centric, but half-echo acquisition is °200 ms faster due to the doubling of y-gradient steps for X-Centric. We believe that our half-echo acquisition vs. X-Centric results indicate that the half-echo approach is 20% faster but 10% less SNR efficient than X-Centric. A structured phantom needs to be used for probing the potential smoothing effects and resolution implications following the Partial Fourier Reconstruction.

#### 3.8 Future role of X-Centric

The focus of this chapter is a presentation of a fast and SNR efficient imaging method for high spatial resolution imaging of mouse lungs that include both the airways and lung parenchyma. Such a technique can be potentially used for mapping the morphological changes and ventilation heterogeneity associated with rodent models of asthma or ovalbumin-challenged model (OVA). Figure 13 shows an example of ventilation heterogeneity maps obtained for a sham mouse (right column) and OVA treated mouse (left column). The figure compares increased (a and c) and decreased (b and d) ventilation heterogeneity indexes from a sham mouse (left column) and an OVA mouse (right column) under similar increases in

#### Figure 12.

Phantom images obtained with partial Fourier reconstruction (PFR, left column) and fast Fourier reconstruction (FFT, right column) for three different matrix sizes are shown. The open source partial Fourier reconstruction (projection onto convex sets (POCS), 2013) was obtained from the official MatLab web site. The following signal-to-noise ratio (SNR) values were obtained for the presented images: SNR = 39 (160 � 256, PFR); SNR = 54 (160 � 256, FFT); SNR = 45 (130 � 256, PFR); and SNR = 58 (256 � 160, FFT).

airways resistance (104% vs. 93%, respectively) and elastance (21% vs. 27%, respectively). The white and red lines in the figure outline the lung contour before and after the Methacholine inhalation (MCh) challenge, respectively. Airway resistance and tissue elastance were measured immediately following each scan. Lung images were then analyzed using local standard deviation (SD) in sliding (9 � 9 pixels<sup>2</sup> ) ROIs. The ventilation heterogeneity index (VHI) was given by:

$$\text{VHI} \left( \% \right) = \frac{\text{SD}\_{\text{post-MCh}} - \text{SD}\_{\text{prc-MCh}}}{\text{SD}\_{\text{prc-MCh}}} \text{x100\%} \tag{15}$$

The post-MCh lung function data were normalized by the pre-MCh values to give the percent change. The sham mouse and an OVA mouse data are shown only as an example of the potential X-Centric pre-clinical application. However, data suggested that ventilation heterogeneity following MCh challenge may be commensurate with traditional lung function testing to access the airway hyperresponsiveness in the OVA mouse model. The observed heterogeneity in ventilation distribution could potentially provide a novel endpoint to study disease modification in asthma, as well as for better diagnosis and classification of asthma in the clinic.

The X-Centric method described here should also be beneficial for clinical hyperpolarized lung MRI and other applications where short-TE techniques are

#### Figure 13.

The VHI maps from a sham mouse (left column) and an OVA treated mouse (right column). The white and red lines in the figure outline the lung contour before and after the Methacholine inhalation challenge, respectively. We applied partial-echo FGRE to acquire a 96 ˜ 48 matrix and zero-filled in phase-encoding direction to get <sup>a</sup> <sup>96</sup> ˜ <sup>96</sup> matrix using FOV <sup>=</sup> <sup>2</sup> ˜ <sup>2</sup> cm<sup>2</sup> , BWread = 83.3 kHz,TE = 0.42 ms and TR = 1.3 ms. the scan time for each image was ° <sup>100</sup> ms and the calculated <sup>b</sup>-value was °0.7 s/cm2 . The mouse was ventilated at 100 breaths/min and 8.5 ml/kg tidal volume using a ventilation pattern consisting of 160 ms of inhalation, 120 ms of breath-hold (for image acquisition) and 320 ms of exhalation. A series of <sup>3</sup> He MR lung images (every 15 s) were acquired before and after methacholine inhalation challenge (160 μg/kg, IV).

needed. Table 2 helps to estimate the acquisition time for BWread = 62.5 kHz and FOV <sup>=</sup> <sup>200</sup> ˜ <sup>200</sup> mm<sup>2</sup> (pediatric study [63]) for two matrix sizes frequently used in human studies [6]: 128 ˜ 128 ˜ 14 slices and 256 ˜ 256 ˜ 14 slices (3D case) yielding resolutions of 3.1 ˜ 3.1 ˜ <sup>15</sup> mm<sup>3</sup> and 1.6 ˜ 1.6 ˜ <sup>15</sup> mm<sup>3</sup> at FOV <sup>=</sup> <sup>400</sup> ˜ <sup>400</sup> ˜ 15mm<sup>3</sup> . With the assumption of partial sampling in the phase-encoding direction and taking into account that the maximum strength of the clinical gradient system is typically 4–5 G/cm with a maximum slew rate of 200 T/m/s [6], one can calculate the acquisition to be around 5 s for a 128 ˜ 128 matrix and 16.7 s for a <sup>256</sup> ˜ <sup>256</sup> matrix. As the <sup>256</sup> ˜ <sup>256</sup> matrix at <sup>a</sup> FOV of <sup>20</sup> ˜ <sup>20</sup> cm2 corresponds to 781 μm nominal resolution, this should lead to 70% signal decay in airways due to diffusion-weighting according to [22] for the case of full-echo FGRE. The X-Centric approach could eliminate 90% of the signal loss due to incidental diffusionweighting in the major airways. Parallel imaging approaches [64] or compressed sensing approaches [65, 66] can also speed up data acquisition even more and are a straight-forward extension of the X-Centric approach described here.

Until now, X-Centric has not been used in any pediatric studies; nevertheless, hyperpolarized gas MRI is gradually becoming an important research tool, allowing

### Figure 14. <sup>3</sup>

He coronal MR images of a 17-month-old non-sedated, unrestrained infant with Cystic Fibrosis. Homogeneous ventilation is seen throughout the lungs. A possible tiny ventilation defect is seen at the right base (arrow). No noticeable motion artifacts are noticeable. 15 spiral interleaved acquisitions (TE = 0.9 ms,TR = 8.1 ms, pixel size <sup>=</sup> <sup>3</sup> ° 3, number of slices <sup>=</sup> 18; slice thickness <sup>=</sup> <sup>10</sup> mm, and FOV <sup>=</sup> <sup>40</sup> ° <sup>40</sup> cm<sup>2</sup> ). Adapted from Altes et al. [28].

for longitudinal observation of lung diseases in children such as Cystic Fibrosis (CF). Figure 14 shows <sup>3</sup> He MRI lung images obtained from a 17-month old patient with CF. Note, the regions of hypo-intensity are difficult to see at this resolution, but using X-Centric, one should be able to gain diagnostic value and better enhance the contrast in these regions. We believe that X-Centric can be very useful for not just CF but the longitudinal observation of other lung diseases like Bronchopulmonary Dysplasia (BPD). Abnormal prematurity at birth (<28 weeks gestation) frequently requires neonatal intensive care and mechanical ventilation due to respiratory distress syndrome due to limited or lack of lung surfactant and incomplete lung development. Ventilator-associated lung injury, along with the extreme prematurity of the lungs themselves, causes abnormal and irregular lung growth (BPD) [67]. The relatively short signal life-time caused by the need to scan newborns inside their neonatal incubators (causing B<sup>0</sup> field distortions, especially at high field) and also the need for smaller FOV make the X-Centric approach ideally suitable for high-resolution neonatal lung MRI.

#### 4. Use of X-centric with other MR-visible nuclei

#### 4.1 Gas phase 129Xe MRI

3 He is a rare and expensive isotope (Table 1), so a worldwide clinical translation of the <sup>3</sup> He MRI method is quite questionable, especially when there is a relatively inexpensive alternative—129Xe MRI. In contrast to the <sup>3</sup> He isotope, the 129Xe isotope is much more abundant (˜26% natural abundance) and cost efficient (Table 1). In addition, recent improvements in the xenon polarization process have enabled sufficient increase in the level of 129Xe polarization (>30%) and production volume

#### High Resolution <sup>3</sup> He Pulmonary MRI DOI: http://dx.doi.org/10.5772/intechopen.84756

of the polarized gas (˜<sup>2</sup> <sup>L</sup> per hour) [68, 69]. Presently, 129Xe lung MRI is translating towards a clinical tool, and it has recently been used as a clinical tool (along with <sup>3</sup> He MRI) in the United Kingdom [70]. The North-American xenon consortium [71] expects 129Xe MRI to be the Food and Drug Administration approved in the first quarter of 2020.

Presently, the main directions of <sup>3</sup> He MRI are pediatric and neonatal studies (using relatively small doses of <sup>3</sup> He) mainly due to the cost of the <sup>3</sup> He isotope and the high quality of the <sup>3</sup> He lung images. In turn, 129Xe MRI is much more suited for adult lung imaging; however, the low SNR of 129Xe images (compared to <sup>3</sup> He) impede the development of many novel acquisition schemes that are highly sensitive to SNR [72], such as isotropic voxel static-ventilation imaging and accelerated multiple b-value diffusion-weighting MRI [73]. Mentioned above, recent improvement in xenon polarization [68, 69] provides a way to develop advanced imaging methods previously not possible. This is especially critical for investigations in patients with asthma and COPD. Here we report the isotropic-voxel, highresolution 3D 129Xe static-ventilation image acquired in a single 16 s breath-hold, with the necessary and sufficient SNR to quantify ventilation defect percent [74] estimates (an efficient asthma and COPD biomarker). To the best of our knowledge, <sup>a</sup> voxel size of <sup>3</sup> ° <sup>3</sup> ° <sup>3</sup> mm<sup>3</sup> is the highest currently achieved resolution for 129Xe human lung images. Figures 15 and 16 show the axial and coronal lung images obtained for a healthy volunteer using an isotropic voxel 3D FGRE pulse sequence. The central slice SNR was around 30 for both axial and coronal view images. The gas phase of 129Xe has a relatively low free diffusion coefficient (0.14 cm<sup>2</sup> /s in air,

#### Figure 15.

Axial view isotropic 129Xe MRI static-ventilation slices from superior to inferior for a healthy volunteer. Isotropic-voxel 129Xe static-ventilation images were acquired using an axial plane 3D fast gradient-recalled Echo sequence (TE/TR/initial-flip-angle = 1.3 ms/4.0 ms/0.9<sup>o</sup> , variable-flip-angle, reconstructed matrix size <sup>=</sup> <sup>128</sup> ° <sup>128</sup> ° <sup>104</sup> [axial view], pixel bandwidth <sup>=</sup> <sup>217</sup> Hz, and FOV <sup>=</sup> <sup>40</sup> ° <sup>40</sup> ° <sup>32</sup> cm<sup>3</sup> , voxel size <sup>=</sup> <sup>3</sup> ° <sup>3</sup> ° <sup>3</sup> mm3 ). 129Xe lung images by and courtesy of G. Parraga.

#### Figure 16.

Coronal view isotropic 129Xe MRI static-ventilation slices from anterior to posterior for a healthy volunteer. Isotropic-voxel 129Xe static-ventilation images were acquired using an axial plane 3D fast gradient-recalled Echo sequence (TE/TR/initial-flip-angle = 1.3 ms/4.0 ms/0.9<sup>o</sup> , variable-flip-angle, reconstructed matrix size <sup>=</sup> <sup>128</sup> ˜ <sup>128</sup> ˜ <sup>104</sup> [axial view], pixel bandwidth <sup>=</sup> <sup>217</sup> Hz, and FOV <sup>=</sup> <sup>40</sup> ˜ <sup>40</sup> ˜ <sup>32</sup> cm<sup>3</sup> , voxel size <sup>=</sup> <sup>3</sup> ˜ <sup>3</sup> ˜ <sup>3</sup> mm3 ). 129Xe lung images by and courtesy of G. Parraga.

Table 1) and relatively long signal life-time (10–30 ms depends on the field strength [75]), so the benefits of the X-Centric approach (overcoming a high diffusivity and short T2\*), may not be very obvious for the case of healthy subjects compared to the traditional FGRE sequence. However, lungs of patients with severe emphysema, COPD (smokers or ex-smokers), BPD and Alpha-1 Antitrypsin Deficiency may show regions with significant terminal airways destruction, changing the diffusion regime to nearly-free diffusion (xenon D ≈ 0.22 cm2 /s when mixed 50/ 50 with helium-4 [6]) in such regions, naturally leading to substantial local reduction of T2\* (up to 50% reduction due to the fast motion through the local B<sup>o</sup> inhomogeneities within the lung airspaces) [73] especially at high field. X-Centric, is a short-TE method that should be able to better visualize lung regions with short T2\*; and therefore, a better tool for helping to make clinical decisions and for therapy planning.

#### 4.2 Dissolved phase 129Xe MRI

Unlike its counterpart <sup>3</sup> He, 129Xe is highly soluble in a variety of solvents and biological materials, and exhibits a large range of chemical shifts within these distinct chemical environments (Ostwald solubility coefficients of 0.17 [40]). Of particular interest is dissolved phase 129Xe residing in and exchanging with lung tissue (tissue barrier) and red blood cells (RBC) [76–78]. The exchange can be measured using MRI and it is also possible distinctly resolve and quantify 129Xe within the

#### High Resolution <sup>3</sup> He Pulmonary MRI DOI: http://dx.doi.org/10.5772/intechopen.84756

tissue barrier and RBC. Exploiting these signals for diagnostic imaging purposes can be difficult since for example, only about 2% of inhaled xenon dissolves into the tissue barrier and RBC [76, 79], and the corresponding T2\* values are often short (˜2 ms) [80]. There is an urgent need for efficient short-TE methods to acquire these short T2\* signals and enable deeper study of a number of lung abnormalities such as ventilation/perfusion mismatch [81, 82] and of other organs including xenon brain perfusion imaging [72, 83] and imaging of the xenon-encapsulated Cryptophane-A functionalized with anti-cancer drugs [84].

The feasibility of the X-centric method was initially demonstrated using <sup>3</sup> He for high-resolution rat lung imaging [39] indicating its ability to overcome diffusion induced signal attenuation within the trachea and substantially reducing TE and t<sup>d</sup> as well as the magnitude of the imaging gradient [27]. The low diffusivity and low gyromagnetic ratio of 129Xe (Table 1) suggest that diffusion decay due to the imaging gradients would not be an obstacle for high-resolution 129Xe gas MRI; however, the relatively short T2\* of the dissolved xenon remains a problem, as it leads to signal decay and contributes to image blurring simultaneously. Clearly, short-TE pulse sequences such as spiral [85, 86] or X-Centric [17, 27] are absolutely necessary to acquire images of xenon dissolved in lung tissue and blood.

Figure 17b shows dissolved phase hyperpolarized xenon obtained from within a water-filled resolution phantom (Figure 17a) on a clinical 3 T MRI scanner (maximum gradient strength was 5 G/cm, sequence parameters are shown in Table 2). The 129Xe image shows that X-Centric can be used to image dissolved phase xenon efficiently (within a breath-hold) and with reasonable signal-to-noise ratio (SNR = 12). After reconstruction, the smallest geometric features of the resolution phantom in the X-Centric image are discernable (yellow arrows). It should be noted, that the available SNR achievable when imaging dissolved phase 129Xe principally depends on concentration, polarization and relaxation times and even physiological parameters such as blood flow and blood oxygenation. The quality of the polarizer, solubility of the solvents used and chemical environments all play a role. Nevertheless, techniques employing significant reductions in echo-times can deliver higher SNR than is typically achievable for spins within environments where their transverse relaxation times are significantly shortened. Though, the phantom dissolved-phase spins had estimated T2\*-values of roughly 20 ms, the potential for using X-Centric for imaging xenon within the tissue barrier, RBC and

#### Figure 17.

Resolution phantom (a) and (b) 129Xe X-centric image obtained in a clinical 3 T MRI scanner. Signal-to-noise ratio value was 12.5. The yellow arrow indicates the 3 mm diameter feature of the resolution phantom demonstrating resolution of the smallest feature of the phantom. The X-centric sequence parameters are shown in Table 2. Adapted from Ouriadov et al. [17].

brain (T2\* < 2 ms) is high and the prospects for high-resolution dissolved phase imaging (currently not available) are demonstrated.

#### 4.3 Static-ventilation 19F MRI

The limited use of the hyperpolarized <sup>3</sup> He/129Xe MRI imaging modality can be partially explained by its need for utilizing expensive isotopes and polarizers, the latter requiring specially-trained personnel. Inert inhaled fluorinated gas MRI can be a promising and less expensive (Table 1) alternative to hyperpolarized <sup>3</sup> He/129Xe MRI. 19F MRI does not require the use of rare isotopes (100% naturally abundant, Table 1) or expensive polarizers. 19F gases (such as perfluoropropane (C3F8) and sulfur hexafluoride (SF6) [17, 18]) can be imaged using high field (≥1.5 T) clinical MRI scanners; however, a multi-nuclear amplifier and receiver along with RF coils tuned to the 19F frequency are still required [18, 19]. Fluorine-19 has a large gyromagnetic ratio (˜95% of <sup>1</sup> H, Table 1), both SF6 and C3F8 have several 19F atoms per molecule, and have relatively short T<sup>1</sup> values making these highly attractive options for researchers (Table 1, [87]). Short T<sup>1</sup> translates to overcoming their low thermal polarization [17, 88] values by rapid signal averaging. Additionally, because it is not hyperpolarized, Fluorine-19 gases can be mixed with O2 without polarization losses, preserving the image quality patient safety simultaneously during a breath-hold scan [89]. Inert fluorinated gases have very short T2\* values (T2\* < 2 ms [87]), therefore imaging methods with short echo times are preferable. As we demonstrated previously, the X-Centric acquisition can be reliably used for phantom and in-vivo measurements using both <sup>3</sup> He and 129Xe gases. Similar to the dissolved 129Xe case, a short apparent transverse relaxation time specific to the fluorinated gas can potentially lead to fast signal attenuation and image blurring. As X-Centric is a short-TE sequence it should be able to ensure sufficient SNR of the fluorinated gas images, and also minimal image blurring. Figure 18a and b show an SF6 high-resolution phantom (Figure 14a) and rat lung images obtained using X-Centric on a clinical 3 T MRI scanner (maximum gradient strength was 5 G/cm, in-vivo image sequence parameters are shown in Table 2). In-vivo data acquisition was synchronized to breath-hold durations using the ventilator [17]. Rat lungs were saturated with an SF6/O2 80/20 gas mixture during 3 min of normal free breathing at a breathing rate of 60 breaths/min. After 3 min, the fluorinated gas/O2 mixture was suspended followed by a 10 sec breath-hold to obtain in-vivo rat lung image (Figure 18b) [17]. Inert fluorinated gas images confirm that X-Centric ensures

#### Figure 18.

(a) An SF6-filled resolution phantom image (FOV <sup>=</sup> <sup>20</sup> ° <sup>20</sup> cm<sup>2</sup> , BWread = 200 Hz/pixel) obtained using Xcentric (TE = 0.5 ms, SNR = 14) and (b) in-vivo 19F X-centric rat lung image obtained in a clinical 3 T MRI scanner. SNR values were 14 and 40 respectively. 19F X-centric sequence parameters are shown in Table 2. Figure adapted from Ouriadov et al. [17].

sufficient SNR (14 and 40 for the phantom and in-vivo images respectively) and resolution (phantom images shows all details) of fluorinated gas images, when apparent transverse relaxation time is less than 2 ms. These results suggest the possibility for translating these techniques directly to human use because they are demonstrated using a clinical scanner using clinical gradient coils.

#### 4.4 Dynamic 19F MRI

19F gas MRI also has a high potential for successful dynamic lung imaging [88, 90] due to the fact that fluorinated gases can be mixed with O2 which helps to restore the initial magnetization faster (decreasing T1), in contrast to the hyperpolarized gases. Furthermore, imaging times are shortened and breath-holds are more easily tolerated by the patients [89]. A technique like this permits the acquisition of regional fractional-ventilation [23, 48, 91] which can be very useful as a radiation free alternative to CT for detecting gas trapping in lung diseases such as lung fibrosis and COPD [88]. Recently, free-breathing 19F (C3F8) dynamic lung imaging has been demonstrated in human lungs [89]. Figure 19 (top panel) shows the gradual wash-out 19F gas within the 19F MRI lung images obtained from a COPD patient for eight wash-out breaths [89]. As expected, each new wash-out breath of air replaced some volume of the fluorinated gas in lung, so the signal intensity of the resulting images was gradually attenuated. The following equation can be fitted to this wash-out data [48]:

$$\mathcal{S}(n) = \mathcal{S}\_0 \cdot (1 - r)^n \tag{16}$$

where S<sup>0</sup> is the initial signal, n is the breath number, S(n) is the signal intensity after a certain number of wash-out breaths and r is the fractional-ventilation <sup>19</sup> parameter [17, 48]. In turn, r can be expressed as the fraction between new F incoming to lung, and the total volume of 19F (Vtotal) [17, 48]:

$$r = \frac{V\_{new}}{V\_{total}} = \frac{V\_{new}}{V\_{old} + V\_{new}} \tag{17}$$

where Vnew and Vold are the "new" and "old" gas volumes in each voxel at each breath. In general, Eq. (16) can be approximated with the stretched exponential (or Kohlrausch) function [92]:

#### Figure 19.

Corresponding 19F gas washout in expiration for one section of the three-dimensional volume (top panel). 19F washout was started at 24.6 s. Images were acquired using a coronal plane 3D spoiled gradient-Echo sequence (TE/TR/flip-angle = 1 ms/3 ms/5o , matrix size = 192 � 130 � 88, pixel bandwidth = 550 Hz, and FOV <sup>=</sup> <sup>50</sup> � 37.5 � 35.2 cm<sup>3</sup> , voxel size <sup>=</sup> 2.6 � 2.9 � <sup>4</sup> mm3 ) with a generalized auto-calibrating partially parallel acquisition with a reduction factor of 2 in a single breath-hold. Figure adapted from Gutberlet et al. [89]. Sparsity pattern for AF = 14 (bottom panel) depicting the k-space under-sampling scheme, ensuring a variety of sparsity patterns for each time point or wash-out breath. AF = accelaration factor.

$$S(n) = \exp\left[\left(-nr\right)^{\beta}\right] \tag{18}$$

where 0 < β ≤ 1, and r is the fractional-ventilation. This simple function has been recently used to combine the CS method with under-sampling in the parametric direction [93], which permitted an acceleration factor (AF) of 7 in multi b-value diffusion-weighted <sup>3</sup> He MRI measurements [73]. Generally, a reasonable choice of the AF value depends on a number of the images in parametric direction. The main idea behind this relationship is to keep the reconstruction errors low [93]. As we stated above in this chapter, X-Centric is Cartesian pulse sequence, which can be naturally used for under-sampling in imaging (phase-encoding) and parametric (time) directions following CS with the stretched exponential method [93]. To our knowledge this is first record of the possibility of using CS combined with the stretched exponential method for dynamic lung imaging studies. Figure 19 (bottom panel) shows an under-sampled X-Centric k-space pattern ensuring AF = 14. Remarkably, this means that only 8% of k-space, or five k-space lines (out of 64) are sampled to achieve a "full dataset" and each image in time can be acquired 13 times faster, which obviously helps to avoid the motion artifacts common to freebreathing data acquisition. Additionally, X-Centric can potentially ensure a much shorter TE compared to the spoiled gradient-echo sequence (TE = 1 ms) used to obtained 19F human lung images [89]. Retrospective X-Centric-like under-sampling of the 19F dynamic rat lung in-vivo data (64 � <sup>64</sup> matrix size, nine wash-out images [17]) available to us has confirmed that AF = 14 led to less than a 12% reconstruction difference (pixel-by-pixel comparison between the fully-sampled image and one reconstructed from the retrospectively under-sampled k-space). Therefore, the choice of X-Centric for dynamic 19F gas MRI lung imaging will speed up imaging time and improve image quality making it an optimal choice for this kind of imaging.

The remaining obstacle preventing wide-spread high-resolution 19F MRI with or without X-Centric is the generally low SNR of 19F lung images. In this particular case, the further improvement of the image quality strongly depends on a hardware component, specifically from RF coil systems. It has been recently demonstrated that the use of a multi-channel phased-receive array can significantly improve the image quality of the 19F MRI lung images [94]. This is a very promising result suggesting that the combination of the X-Centric method with advanced array-type RF coil system will permit future high-resolution 19F MRI human lung imaging in both static-ventilation and dynamic studies.

#### 5. Conclusion

In summary, hyperpolarized <sup>3</sup> He X-Centric MR imaging with high spatial resolution was proven to be a robust technique in phantom and mouse lung measurements yielding a nominal resolution of 39 μm and 78 μm respectively. In particular, mouse major airways with less restricted diffusion of <sup>3</sup> He (<sup>D</sup> � 0.9 cm2 /s) could only be visualized with the X-Centric method. Note, that the high nominal resolution (78 μm) in in-vivo scans was achieved within 1 s breath-hold. These results suggest that the X-Centric method can potentially fill a gap and need for high spatial and temporal resolution imaging method for the small model of asthma studies.

Another beneficial feature of the X-Centric sequence is the significantly reduced echo-time, so the method can be used for imaging of nuclei with short signal lifetimes such as hyperpolarized 129Xe dissolved within the lung tissue barrier and/or

High Resolution <sup>3</sup> He Pulmonary MRI DOI: http://dx.doi.org/10.5772/intechopen.84756

Red Blood Cells, and inert fluorinated gas (SF6) inside the lungs. In both cases the T2\* value is less than 2 ms. The feasibility and future potential of short-TE X-Centric-based X-Nuclei measurements were demonstrated using 129Xe dissolved in water as well as SF6 within rat lungs.

#### Acknowledgements

The authors would like to thank the following individuals for assistance with MRI experiments and data analysis: Grace E. Parraga, Giles E. Santyr, Mitchell S. Albert, Marcus J. Couch, Tao Li and Iain K. Ball. A special thanks to Ben T. Chen for providing support for the hyperpolarized <sup>3</sup> He gas, animals, and ventilator.

#### Conflict of interest

The authors do not have any conflict of interest.

#### Notes/thanks/other declarations

We thank Michael Völker for providing the MatLab code (Projection onto Convex Sets (POCS), 2013) for image reconstruction. We thank Abascal et al. for providing the MatLab code (Signal Decay Into the Reconstruction (SIDER), 2017) for image reconstruction.

#### Nomenclature



### Author details

Matthew S. Fox1,2 and Alexei V. Ouriadov1,2\*

1 Lawson Health Research Institute, London, ON, Canada

2 Department of Physics and Astronomy, Western University, London, ON, Canada

\*Address all correspondence to: aouriado@uwo.ca

© 2019 The Author(s). Licensee IntechOpen. This chapteris distributed underthe terms oftheCreative Commons Attribution License (http://creativecommons.org/licenses/ by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

High Resolution <sup>3</sup> He Pulmonary MRI DOI: http://dx.doi.org/10.5772/intechopen.84756

#### References

[1] Albert MS, Cates GD, Driehuys B, Happer W, Saam B, Springer CS, et al. Biological magnetic resonance imaging using laser-polarized 129Xe. Nature. 1994;370(6486):199-201

[2] Saam BT, Yablonskiy DA, Kodibagkar VD, Leawoods JC, Gierada DS, Cooper JD, et al. MR imaging of diffusion of 3He gas in healthy and diseased lungs. Magnetic Resonance in Medicine. 2000;44(2):174-179

[3] Salerno M, de Lange EE, Altes TA, Truwit JD, Brookeman JR, Mugler JP 3rd. Emphysema: Hyperpolarized helium 3 diffusion MR imaging of the lungs compared with spirometric indexes—Initial experience. Radiology. 2002;222(1):252-260

[4] Swift AJ, Wild JM, Fichele S, Woodhouse N, Fleming S, Waterhouse J, et al. Emphysematous changes and normal variation in smokers and COPD patients using diffusion 3He MRI. European Journal of Radiology. 2005; 54(3):352-358. DOI: 10.1016/j. ejrad.2004.08.002

[5] Evans A, McCormack D, Ouriadov A, Etemad-Rezai R, Santyr G, Parraga G. Anatomical distribution of 3He apparent diffusion coefficients in severe chronic obstructive pulmonary disease. Journal of Magnetic Resonance Imaging. 2007; 26(6):1537-1547. DOI: 10.1002/ jmri.21205

[6] Kirby M, Svenningsen S, Owrangi A, Wheatley A, Farag A, Ouriadov A, et al. Hyperpolarized 3He and 129Xe MR imaging in healthy volunteers and patients with chronic obstructive pulmonary disease. Radiology. 2012; 265(2):600-610

[7] Altes TA, Powers PL, Knight-Scott J, Rakes G, Platts-Mills TA, de Lange EE, et al. Hyperpolarized 3He MR lung ventilation imaging in asthmatics:

Preliminary findings. Journal of Magnetic Resonance Imaging. 2001; 13(3):378-384. DOI: 10.1002/jmri.1054

[8] de Lange EE, Altes TA, Patrie JT, Gaare JD, Knake JJ, Mugler JP 3rd, et al. Evaluation of asthma with hyperpolarized helium-3 MRI: Correlation with clinical severity and spirometry. Chest. 2006;130(4). DOI: 1055-62. DOI 10.1378/chest.130.4.1055

[9] Tzeng YS, Lutchen K, Albert M. The difference in ventilation heterogeneity between asthmatic and healthy subjects quantified using hyperpolarized 3He MRI. Journal of Applied Physiology (Bethesda, MD: 1985). 2009;106(3): 813-822. DOI: 10.1152/japplphysiol. 01133.2007

[10] Fain SB, Gonzalez-Fernandez G, Peterson ET, Evans MD, Sorkness RL, Jarjour NN, et al. Evaluation of structure-function relationships in asthma using multidetector CT and hyperpolarized He-3 MRI. Academic Radiology. 2008;15(6):753-762. DOI: 10.1016/j.acra.2007.10.019

[11] Svenningsen S, Eddy RL, Lim HF, Cox PG, Nair P, Parraga G. Sputum eosinophilia and magnetic resonance imaging ventilation heterogeneity in severe asthma. American Journal of Respiratory and Critical Care Medicine. 2018;197(7):876-884. DOI: 10.1164/ rccm.201709-1948OC

[12] Hoover DA, Capaldi DP, Sheikh K, Palma DA, Rodrigues GB, Dar AR, et al. Functional lung avoidance for individualized radiotherapy (FLAIR): Study protocol for a randomized, double-blind clinical trial. BMC Cancer. 2014;14(1):934. DOI: 10.1186/ 1471-2407-14-934

[13] Ontario Lung Association. Your Lungs, Your Life: Insights and Solutions to Lung Health in Ontario. 2012:1-24

[14] Canadian Cancer Society. Special topic: Pancreatic cancer. Canadian Cancer Statistics 2017. 2017:1-142

[15] Kovtunov KV, Truong ML, Barskiy DA, Koptyug IV, Coffey AM, Waddell KW, et al. Long-lived spin states for low-field hyperpolarized gas MRI. Chemistry. 2014;20(45):14629-14632. DOI: 10.1002/chem.201405063

[16] Kovtunov KV, Romanov AS, Salnikov OG, Barskiy DA, Chekmenev EY, Koptyug IV. Gas phase UTE MRI of propane and propene. Tomography. 2016;2(1):49-55. DOI: 10.18383/j. tom.2016.00112

[17] Ouriadov AV, Fox MS, Couch MJ, Li T, Ball IK, Albert MS. In vivo regional ventilation mapping using fluorinated gas MRI with an x-centric FGRE method. Magnetic Resonance in Medicine. 2015;74(2):550-557. DOI: 10.1002/mrm.25406

[18] Couch MJ, Ball IK, Li T, Fox MS, Littlefield SL, Biman B, et al. Pulmonary ultrashort echo time 19F MR imaging with inhaled fluorinated gas mixtures in healthy volunteers: Feasibility. Radiology. 2013;269(3):903-909. DOI: 10.1148/radiol.13130609

[19] Halaweish AF, Moon RE, Foster WM, Soher BJ, McAdams HP, MacFall JR, et al. Perfluoropropane gas as a magnetic resonance lung imaging contrast agent in humans. Chest. 2013; 144(4):1300-1310. DOI: 10.1378/ chest.12-2597

[20] Wang W, Nguyen NM, Yablonskiy DA, Sukstanskii AL, Osmanagic E, Atkinson JJ, et al. Imaging lung microstructure in mice with hyperpolarized 3He diffusion MRI. Magnetic Resonance in Medicine. 2011; 65(3):620-626. DOI: 10.1002/ mrm.22737

[21] Walvick RP, Roche JP, Reno AL, Gounis MJ, Albert MS. Evaluation of oxygen sensitivity of hyperpolarized helium imaging for the detection of pulmonary ischemia. Magnetic Resonance in Medicine. 2016;75(5): 2050-2054. DOI: 10.1002/mrm.25714

[22] Driehuys B, Walker J, Pollaro J, Cofer GP, Mistry N, Schwartz D, et al. 3He MRI in mouse models of asthma. Magnetic Resonance in Medicine. 2007; 58(5):893-900. DOI: 10.1002/mrm. 21306

[23] Couch MJ, Ouriadov A, Santyr GE. Regional ventilation mapping of the rat lung using hyperpolarized 129Xe magnetic resonance imaging. Magnetic Resonance in Medicine. 2012;68(5): 1623-1631. DOI: 10.1002/mrm.24152

[24] Wild JM, Woodhouse N, Paley MN, Fichele S, Said Z, Kasuboski L, et al. Comparison between 2D and 3D gradient-echo sequences for MRI of human lung ventilation with hyperpolarized 3He. Magnetic Resonance in Medicine. 2004;52(3): 673-678. DOI: 10.1002/mrm.20164

[25] Dugas JP, Garbow JR, Kobayashi DK, Conradi MS. Hyperpolarized (3)He MRI of mouse lung. Magnetic Resonance in Medicine. 2004;52(6): 1310-1317. DOI: 10.1002/mrm.20300

[26] Mugler JP, Brookeman JR. Method and System for Rapid Magnetic Resonance Imaging of Gases with Reduced Diffusion-Induced Signal Loss. Google Patents; 2006

[27] Ouriadov AV, Santyr GE. High spatial resolution hyperpolarized (3) He MRI of the rodent lung using a single breath X-centric gradient-recalled echo approach. Magnetic Resonance in Medicine. 2017;78(6):2334-2341. DOI: 10.1002/mrm.26602

[28] Altes TA, Meyer CH, Mata JF, Froh DK, Paget-Brown A, Gerald Teague W, et al. Hyperpolarized helium-3 magnetic resonance lung imaging of non-sedated

High Resolution <sup>3</sup> He Pulmonary MRI DOI: http://dx.doi.org/10.5772/intechopen.84756

infants and young children: A proof-ofconcept study. Clinical Imaging. 2017; 45:105-110. DOI: 10.1016/j. clinimag.2017.04.004

[29] Bergin CJ, Pauly JM, Macovski A. Lung parenchyma: Projection reconstruction MR imaging. Radiology. 1991;179(3):777-781. DOI: 10.1148/ radiology.179.3.2027991

[30] Chen BT, Brau AC, Johnson GA. Measurement of regional lung function in rats using hyperpolarized 3helium dynamic MRI. Magnetic Resonance in Medicine. 2003;49(1):78-88. DOI: 10.1002/mrm.10336

[31] Viallon M, Berthezene Y, Callot V, Bourgeois M, Humblot H, Briguet A, et al. Dynamic imaging of hyperpolarized3He distribution in rat lungs using interleaved-spiral scans. NMR in Biomedicine. 2000;13(4): 207-213. DOI: 10.1002/1099-1492 (200006)13:4<207::aid-nbm641>3.0. co;2-g

[32] Salerno M, Altes TA, Brookeman JR, de Lange EE, Mugler JP 3rd. Dynamic spiral MRI of pulmonary gas flow using hyperpolarized (3)He: Preliminary studies in healthy and diseased lungs. Magnetic Resonance in Medicine. 2001; 46(4):667-677

[33] Gurney PT, Hargreaves BA, Nishimura DG. Design and analysis of a practical 3D cones trajectory. Magnetic Resonance in Medicine. 2006;55(3): 575-582. DOI: 10.1002/mrm.20796

[34] Wang JX, Ouriadov A, Stainsby J, Parraga G, Santyr G. 3D hyperpolarized 3He imaging of human lung with isotropic spatial resolution using 3Dcones. In: Proceedings of the 17th Annual Meeting of ISMRM; Stockholm, Sweden. 2009. p. 4601

[35] Zhang Z, Ouriadov AV, Willson C, Balcom BJ. Membrane gas diffusion measurements with MRI. Journal of

Magnetic Resonance. 2005;176(2): 215-222. DOI: 10.1016/j. jmr.2005.06.009

[36] Shattuck MD, Gewalt SL, Glover GH, Hedlund LW, Johnson GA. MR microimaging of the lung using volume projection encoding. Magnetic Resonance in Medicine. 1997;38(6): 938-942

[37] Fox MS, Ouriadov A, Santyr GE. Comparison of hyperpolarized 3He and 129Xe MRI for the measurement of absolute ventilated lung volume in rats. Magnetic Resonance in Medicine. 2014; 71(3):1130-1136. DOI: 10.1002/ mrm.24746

[38] Ouriadov AV, Kennan R, Slipetz D, Santyr G, Williams D, Rutt BK, et al. Invivo hyperpolarized 3He lung imaging in mice using x-centric fGRE sequence and custom-designed flexivent ventilator. In: Proceedings of the 16th Annual Meeting of ISMRM; Toronto, Canada. 2008. p. 3789

[39] Santyr GE, Lam WW, Parra-Robles JM, Taves TM, Ouriadov AV. Hyperpolarized noble gas magnetic resonance imaging of the animal lung: Approaches and applications. Journal of Applied Physics. 2009;105(10):102004. DOI: 10.1063/1.3112143

[40] Moller HE, Chen XJ, Saam B, Hagspiel KD, Johnson GA, Altes TA, et al. MRI of the lungs using hyperpolarized noble gases. Magnetic Resonance in Medicine. 2002; 47(6):1029-1051. DOI: 10.1002/ mrm.10173

[41] Mugler JP 3rd, Altes TA. Hyperpolarized 129Xe MRI of the human lung. Journal of Magnetic Resonance Imaging. 2013;37(2):313-331. DOI: 10.1002/jmri.23844

[42] Brown RW, Cheng YCN, Haacke EM, Thompson MR, Venkatesan R. Magnetic Resonance Imaging: Physical Principles and Sequence Design. 2nd ed. Hoboken, New Jersey: Wiley Blackwell; 2014. DOI: 10.1002/978111 8633953

[43] Foster-Gareau P, Heyn C, Alejski A, Rutt BK. Imaging single mammalian cells with a 1.5 T clinical MRI scanner. Magnetic Resonance in Medicine. 2003; 49(5):968-971. DOI: 10.1002/ mrm.10417

[44] Chen XJ, Möller HE, Chawla MS, Cofer GP, Driehuys B, Hedlund LW, et al. Spatially resolved measurements of hyperpolarized gas properties in the lung in vivo. Part II: T∗2. Magnetic Resonance in Medicine. 1999;42(4): 729-737. DOI: 10.1002/(SICI)1522-2594 (199910)42:4<729::AID-MRM15>3.0. CO;2-2

[45] Saam B, Happer W, Middleton H. Nuclear relaxation of <sup>3</sup> He in the presence of O2. Physical Review A. 1995; 52(1):862-865

[46] Deninger AJ, Eberle B, Ebert M, Grossmann T, Heil W, Kauczor HU, et al. Quantification of regional intrapulmonary oxygen partial pressure evolution during apnea by He-3 MRI. Journal of Magnetic Resonance. 1999; 141(2):207-216. DOI: 10.1006/ jmre.1999.1902

[47] Ouriadov AV, Lam WW, Santyr GE. Rapid 3-D mapping of hyperpolarized 3He spin-lattice relaxation times using variable flip angle gradient echo imaging with application to alveolar oxygen partial pressure measurement in rat lungs. Magma. 2009;22(5):309-318. DOI: 10.1007/ s10334-009-0181-3

[48] Santyr GE, Lam WW, Ouriadov A. Rapid and efficient mapping of regional ventilation in the rat lung using hyperpolarized 3He with flip angle variation for offset of RF and relaxation (FAVOR). Magnetic Resonance in Medicine. 2008;59(6):1304-1310. DOI: 10.1002/mrm.21582

[49] Nishimura DG. Principles of Magnetic Resonance Imaging. Stanford University; 1996

[50] Chen XJ, Moller HE, Chawla MS, Cofer GP, Driehuys B, Hedlund LW, et al. Spatially resolved measurements of hyperpolarized gas properties in the lung in vivo. Part I: Diffusion coefficient. Magnetic Resonance in Medicine. 1999;42(4):721-728. DOI: 10.1002/(SICI)1522-2594(199910)42:4 <721::AID-MRM14>3.0.CO;2-D

[51] Miller G, Altes T, Brookeman J, De Lange E, Mugler Iii J. Hyperpolarized 3He lung ventilation imaging with B 1 inhomogeneity correction in a single breath-hold scan. Magnetic Resonance Materials in Physics, Biology and Medicine. 2004;16(5):218-226

[52] Zhao L, Mulkern R, Tseng CH, Williamson D, Patz S, Kraft R, et al. Gradient-echo imaging considerations for hyperpolarized 129Xe MR. Journal of Magnetic Resonance—Series B. 1996; 113(2):179-183

[53] Stejskal EO, Tanner JE. Spin diffusion measurements: Spin echoes in the presence of a time-dependent field gradient. The Journal of Chemical Physics. 1965;42(1):288-292

[54] Danneman PJ, Suckow MA, Brayton C. The Laboratory Mouse. CRC Press; 2012

[55] Chen BT, Yordanov AT, Johnson GA. Ventilation-synchronous magnetic resonance microscopy of pulmonary structure and ventilation in mice. Magnetic Resonance in Medicine. 2005;53(1):69-75. DOI: 10.1002/ mrm.20307

[56] Lundblad L, Thompson-Figueroa J, Leclair T, G Irvin C, Bates J. Thoracic gas volume measurements in paralyzed mice. Annals of Biomedical Engineering. 2004;32(10):1420-1427. DOI: 10.1114/B: ABME.0000042229.41098.6a

High Resolution <sup>3</sup> He Pulmonary MRI DOI: http://dx.doi.org/10.5772/intechopen.84756

[57] Han H, MacGregor RP, Balcom BJ. Pure phase encode magnetic field gradient monitor. Journal of Magnetic Resonance. 2009;201(2):212-217. DOI: 10.1016/j.jmr.2009.09.011

[58] Dominguez-Viqueira W, Ouriadov A, O'Halloran R, Fain SB, Santyr GE. Signal-to-noise ratio for hyperpolarized (3)He MR imaging of human lungs: A 1.5 T and 3 T comparison. Magnetic Resonance in Medicine. 2011;66(5): 1400-1404. DOI: 10.1002/mrm.22920

[59] Rose A. The sensitivity performance of the human eye on an absolute scale. Journal of the Optical Society of America. 1948;38(2):196-208

[60] Haacke EM, Brown RW, Thompson MR, Venkatesan R. Magnetic Resonance Imaging: Physical Principles and Sequence Design. Magnetic Resonance Imaging: Physical Principles and Sequence Design. New York: NY John Wiley & Sons Ltd; 1999. DOI: 10.1002/ 9781118633953.ch1

[61] Stenger VA, Noll DC, Boada FE. Partial Fourier reconstruction for threedimensional gradient echo functional MRI: Comparison of phase correction methods. Magnetic Resonance in Medicine. 1998;40(3):481-490

[62] Haacke EM, Lindskogj ED, Lin W. A fast, iterative, partial-fourier technique capable of local phase recovery. Journal of Magnetic Resonance (1969). 1991; 92(1):126-145. DOI: 10.1016/0022-2364 (91)90253-p

[63] Altes TA, Meyer CH, Mata JH, Froh D, Paget-Brown A, Fain S, et al. Hyperpolarized Helium-3 MR imaging of a non-sedated infant: A proof-ofconcept study: Preliminary experience. In: Proceedings of the 20th Annual Meeting of ISMRM. Melbourne, Australia; 2012. p. P1355

[64] Chang YV, Quirk JD, Yablonskiy DA. In vivo lung morphometry with

accelerated hyperpolarized (3) He diffusion MRI: A preliminary study. Magnetic Resonance in Medicine. 2015; 73(4):1609-1614. DOI: 10.1002/ mrm.25284

[65] Chan HF, Stewart NJ, Parra-Robles J, Collier GJ, Wild JM. Whole lung morphometry with 3D multiple b-value hyperpolarized gas MRI and compressed sensing. Magnetic Resonance in Medicine. 2017;77(5):1916-1925. DOI: 10.1002/mrm.26279

[66] Zhang H, Xie J, Xiao S, Zhao X, Zhang M, Shi L, et al. Lung morphometry using hyperpolarized (129) Xe multi-b diffusion MRI with compressed sensing in healthy subjects and patients with COPD. Medical Physics. 2018;45(7):3097-3108. DOI: 10.1002/mp.12944

[67] Westcott A, McCormack DG, Parraga G, Ouriadov A. Advanced pulmonary MRI to quantify alveolar and acinar duct abnormalities: Current status and future clinical applications. Journal of Magnetic Resonance Imaging. 2019;(0):0. DOI: 10.1002/nbm.3165

[68] Norquay G, Collier GJ, Rao M, Stewart NJ, Wild JM. 129Xe-Rb spinexchange optical pumping with high photon efficiency. Physical Review Letters. 2018;121(15): 153201. DOI: 10.1103/ PhysRevLett.121.153201

[69] Goodson BM, Ranta K, Skinner JG, Coffey AM, Nikolaou P, Gemeinhardt M, et al. Chapter 2—The physics of hyperpolarized gas MRI A2. In: Albert MS, Hane FT, editors. Hyperpolarized and Inert Gas MRI. Boston: Academic Press; 2017. pp. 23-46. DOI: 10.1016/ B978-0-12-803675-4.00002-6

[70] Wild JM, Collier G, Marshall H, Smith L, Norquay G, Swift AJ, et al. P283 hyperpolarised gas MRI—A pathway to Clinical diagnostic imaging. Thorax. 2015;70(Supp. 3):

A220.3-A220A1. DOI: 10.1136/ thoraxjnl-2015-207770.419

[71] 129Xe MRI Clinical Trials Consortium. [Internet]. 2018. https://c pircchmcorg/XeMRICTC.

[72] Rao MR, Stewart NJ, Griffiths PD, Norquay G, Wild JM. Imaging human brain perfusion with inhaled hyperpolarized 129Xe MR imaging. Radiology. 2018;286(2):659-665. DOI: 10.1148/radiol.2017162881

[73] Westcott A, Guo F, Parraga G, Ouriadov A. Rapid single-breath hyperpolarized Noble gas MRI based biomarkers of airspace enlargement. Journal of Magnetic Resonance Imaging. 2018;2018. DOI: 10.1002/ jmri.26574

[74] Svenningsen S, Kirby M, Starr D, Leary D, Wheatley A, Maksym GN, et al. Hyperpolarized (3) He and (129) Xe MRI: Differences in asthma before bronchodilation. Journal of Magnetic Resonance Imaging. 2013; 38(6):1521-1530. DOI: 10.1002/ jmri.24111

[75] Xu X, Norquay G, Parnell SR, Deppe MH, Ajraoui S, Hashoian R, et al. Hyperpolarized 129Xe gas lung MRI-SNR and T2\* comparisons at 1.5 T and 3 T. Magnetic Resonance in Medicine. 2012;68(6):1900-1904. DOI: 10.1002/ mrm.24190

[76] Mugler JP, Driehuys B, Brookeman JR, Cates GD, Berr SS, Bryant RG, et al. MR imaging and spectroscopy using hyperpolarized 129Xe gas: Preliminary human results. Magnetic Resonance in Medicine. 1997;37(6):809-815. DOI: 10.1002/mrm.1910370602

[77] Sakai K, Bilek AM, Oteiza E, Walsworth RL, Balamore D, Jolesz FA, et al. Temporal dynamics of hyperpolarized 129Xe resonances in living rats. Journal of Magnetic Resonance. Series B. 1996;111(3): 300-304

[78] Ruppert K, Brookeman JR, Hagspiel KD, Driehuys B, Mugler JP. NMR of hyperpolarized129Xe in the canine chest: Spectral dynamics during a breath-hold. NMR in Biomedicine. 2000;13(4):220-228. DOI: 10.1002/ 1099-1492(200006)13:4<220::aidnbm638>3.0.co;2-f

[79] Kaushik SS, Robertson SH, Freeman MS, He M, Kelly KT, Roos JE, et al. Single-breath clinical imaging of hyperpolarized (129)Xe in the airspaces, barrier, and red blood cells using an interleaved 3D radial 1-point Dixon acquisition. Magnetic Resonance in Medicine. 2016;75(4):1434-1443. DOI: 10.1002/mrm.25675

[80] Driehuys B, Cofer GP, Pollaro J, Mackel JB, Hedlund LW, Johnson GA. Imaging alveolar–capillary gas transfer using hyperpolarized 129Xe MRI. Proceedings of the National Academy of Sciences. 2006;103(48):18278-18283. DOI: 10.1073/pnas.0608458103

[81] Fox MS, Ouriadov A, Thind K, Hegarty E, Wong E, Hope A, et al. Detection of radiation induced lung injury in rats using dynamic hyperpolarized (129)Xe magnetic resonance spectroscopy. Medical Physics. 2014;41(7):072302. DOI: 10.1118/1.4881523

[82] Doganay O, Matin T, Chen M, Kim M, McIntyre A, McGowan DR, et al. Time-series hyperpolarized xenon-129 MRI of lobar lung ventilation of COPD in comparison to V/Q-SPECT/CT and CT. European Radiology. 2018. DOI: 10.1007/s00330-018-5888-y

[83] Mazzanti ML, Walvick RP, Zhou X, Sun Y, Shah N, Mansour J, et al. Distribution of hyperpolarized xenon in the brain following sensory stimulation: Preliminary MRI findings. PLoS One. 2011;6(7):e21607. DOI: 10.1371/journal. pone.0021607

[84] Zeng Q, Guo Q, Yuan Y, Yang Y, Zhang B, Ren L, et al. Mitochondria

High Resolution <sup>3</sup> He Pulmonary MRI DOI: http://dx.doi.org/10.5772/intechopen.84756

targeted and intracellular biothiol triggered hyperpolarized (129)Xe magnetofluorescent biosensor. Analytical Chemistry. 2017;89(4): 2288-2295. DOI: 10.1021/acs. analchem.6b03742

[85] Zanette B, Stirrat E, Jelveh S, Hope A, Santyr G. Physiological gas exchange mapping of hyperpolarized (129) Xe using spiral-IDEAL and MOXE in a model of regional radiation-induced lung injury. Medical Physics. 2018; 45(2):803-816. DOI: 10.1002/mp.12730

[86] Doganay O, Wade T, Hegarty E, McKenzie C, Schulte RF, Santyr GE. Hyperpolarized (129) Xe imaging of the rat lung using spiral IDEAL. Magnetic Resonance in Medicine. 2016;76(2): 566-576. DOI: 10.1002/mrm.25911

[87] Chang YV, Conradi MS. Relaxation and diffusion of perfluorocarbon gas mixtures with oxygen for lung MRI. Journal of Magnetic Resonance. 2006; 181(2):191-198. DOI: 10.1016/j. jmr.2006.04.003

[88] Couch MJ, Fox MS, Viel C, Gajawada G, Li T, Ouriadov AV, et al. Fractional ventilation mapping using inert fluorinated gas MRI in rat models of inflammation and fibrosis. NMR in Biomedicine. 2016;29(5):545-552. DOI: 10.1002/nbm.3493

[89] Gutberlet M, Kaireit TF, Voskrebenzev A, Lasch F, Freise J, Welte T, et al. Free-breathing dynamic (19)F gas MR imaging for mapping of regional lung ventilation in patients with COPD. Radiology. 2018;286(3): 1040-1051. DOI: 10.1148/ radiol.2017170591

[90] Couch MJ, Ouriadov AV, Albert MS. Chapter 18—Pulmonary imaging using 19f mri of inert fluorinated gases. In: Hyperpolarized and Inert Gas MRI. Boston: Academic Press; 2017. pp. 279-292. DOI: 10.1016/B978-0- 12-803675-4.00018-X

[91] Deninger AJ, Månsson S, Petersson JS, Pettersson G, Magnusson P, Svensson J, et al. Quantitative measurement of regional lung ventilation using 3He MRI. Magnetic Resonance in Medicine. 2002;48(2): 223-232. DOI: 10.1002/mrm.10206

[92] Berberan-Santos MN, Bodunov EN, Valeur B. Mathematical functions for the analysis of luminescence decays with underlying distributions 1. Kohlrausch decay function (stretched exponential). Chemical Physics. 2005; 315(1–2):171-182. DOI: 10.1016/j. chemphys.2005.04.006

[93] Abascal JFPJ, Desco M, Parra-Robles J. Incorporation of Prior Knowledge of the Signal Behavior into the Reconstruction to Accelerate the Acquisition of MR Diffusion Data. ArXiv e-prints [Internet]. 2017; 1702. Available from: http://adsabs.harvard.ed u/abs/2017arXiv170202743A

[94] Charles C, Jones2 RW, Halaweish AF, Ainslie MD. Parallel imaging for short breath hold times in perfluorinated gas imaging of the lung [abstract]. In: Proceedings of the 23rd Annual Meeting of ISMRM; Toronto, Canada. 2015. p. 3984

**87**

Section 5

Uses of MRI in

Radiotherapy

Section 5

Uses of MRI in Radiotherapy

**89**

**Chapter 6**

**Abstract**

4D-MRI in radiotherapy.

motion artifacts

**1. Introduction**

4D-MRI in Radiotherapy

Four-dimensional (4D) imaging provides a useful estimation of tissue motion pattern and range for radiation therapy of moving targets. 4D-CT imaging has been a standard care of practice for stereotactic body radiation therapy of moving targets. Recently, 4D-MRI has become an emerging developmental area in radiotherapy. In comparison with 4D-CT imaging, 4D-MRI provides better spatial rendering of radiotherapy targets in abdominal and pelvis regions with improved visualization of soft tissue motion. Successful implementation of 4D-MRI requires an integration of optimized acquisition protocols, advanced image reconstruction techniques, and sufficient hardware capabilities. The proposed chapter intends to introduce basic theories, current research, development, and applications of

**Keywords:** 4D-MRI, radiotherapy, image reconstruction, respiratory motion,

The role of modern radiotherapy in cancer treatment is to irradiate target volumes that contain disease sites while sparing surrounding normal tissue. In classic 3D-based radiotherapy, treatment volumes are typically defined in **Figure 1(a)**. Gross tumor volume (GTV) contains the primary tumor or malignant tissue, and clinical target volume (CTV) contains GTV plus its surrounding tissue that may have subclinical disease that cannot be definitely revealed by medical imaging (though it is uncommon that CTV is identical to GTV in certain definitive radiotherapy). Planning target volume (PTV), which is often defined as treatment volume in a radiotherapy plan, is defined as CTV plus a margin that accounts for possible tissue displacement and patient positioning uncertainty within a treatment course that may last several weeks [1, 2]. This CTV to PTV margin can be called as setup margin. Depending on different treatment sites and disease stage, setup margin ranges from a few millimeters to 1–2 cm. However, in some radiotherapy treatment, the GTV volume is not stable: when treating tumors in lung, esophagus, and abdominal regions (liver, pancreas, etc.), tumors move under the effect of respiratory activity. Such target motion, referred as respiratory motion, has to be accounted in radiotherapy for effective therapeutic outcome. Thus, a large setup margin for PTV was proposed to account for possible respiratory motion. However, such simple solution has two problems: (1) amplitudes of respiratory motion vary among different individuals. Results of 1–2 mm up to 3 cm are commonly observed in clinic. A single large margin may not yield optimal treatment outcome for patients with extended/limited motion amplitudes [3]; and (2) a large setup margin may lead to unnecessary irradiation of normal tissue, which may substantially increase toxicity of radiotherapy. With the

*Chunhao Wang and Fang-Fang Yin*

## **Chapter 6**  4D-MRI in Radiotherapy

*Chunhao Wang and Fang-Fang Yin* 

### **Abstract**

Four-dimensional (4D) imaging provides a useful estimation of tissue motion pattern and range for radiation therapy of moving targets. 4D-CT imaging has been a standard care of practice for stereotactic body radiation therapy of moving targets. Recently, 4D-MRI has become an emerging developmental area in radiotherapy. In comparison with 4D-CT imaging, 4D-MRI provides better spatial rendering of radiotherapy targets in abdominal and pelvis regions with improved visualization of soft tissue motion. Successful implementation of 4D-MRI requires an integration of optimized acquisition protocols, advanced image reconstruction techniques, and sufficient hardware capabilities. The proposed chapter intends to introduce basic theories, current research, development, and applications of 4D-MRI in radiotherapy.

**Keywords:** 4D-MRI, radiotherapy, image reconstruction, respiratory motion, motion artifacts

#### **1. Introduction**

 The role of modern radiotherapy in cancer treatment is to irradiate target volumes that contain disease sites while sparing surrounding normal tissue. In classic 3D-based radiotherapy, treatment volumes are typically defined in **Figure 1(a)**. Gross tumor volume (GTV) contains the primary tumor or malignant tissue, and clinical target volume (CTV) contains GTV plus its surrounding tissue that may have subclinical disease that cannot be definitely revealed by medical imaging (though it is uncommon that CTV is identical to GTV in certain definitive radiotherapy). Planning target volume (PTV), which is often defined as treatment volume in a radiotherapy plan, is defined as CTV plus a margin that accounts for possible tissue displacement and patient positioning uncertainty within a treatment course that may last several weeks [1, 2]. This CTV to PTV margin can be called as setup margin. Depending on different treatment sites and disease stage, setup margin ranges from a few millimeters to 1–2 cm. However, in some radiotherapy treatment, the GTV volume is not stable: when treating tumors in lung, esophagus, and abdominal regions (liver, pancreas, etc.), tumors move under the effect of respiratory activity. Such target motion, referred as respiratory motion, has to be accounted in radiotherapy for effective therapeutic outcome. Thus, a large setup margin for PTV was proposed to account for possible respiratory motion. However, such simple solution has two problems: (1) amplitudes of respiratory motion vary among different individuals. Results of 1–2 mm up to 3 cm are commonly observed in clinic. A single large margin may not yield optimal treatment outcome for patients with extended/limited motion amplitudes [3]; and (2) a large setup margin may lead to unnecessary irradiation of normal tissue, which may substantially increase toxicity of radiotherapy. With the

**Figure 1.** 

*Diagrams of volume definition in modern radiotherapy. (a) Definitions without consideration of target motion; (b) Definitions with ITV included to account for target motion.* 

 current trend of stereotactic body radiotherapy (SBRT, also referred as stereotactic ablative body radiotherapy SABR) using very high radiation dose to treatment cancer in a few fractions, such big margin with large toxicity becomes unacceptable.

 To account for aforementioned issue, dynamic imaging concept was proposed to capture individualized target motion pattern and amplitude. Such information can be used to define an internal target volume (ITV), which adds an internal margin to CTV that accounts for full possible motion range during radiotherapy (**Figure 1(b)**). This internal margin is determined on an individual basis during the initial treatment simulation. Compared to a generous setup margin, the added internal margin can maximize the therapeutic effect while reducing the irradiation to normal tissue [4, 5].

 Prior to 4D imaging in radiotherapy, X-ray fluoroscopy imaging using C-arm device was the early effort to determine individualized internal margin [6]. The lack of volumetric information in this approach cannot capture the potential motion pattern heterogeneity. Since CT is the dominant modality for radiotherapy with its irreplaceable tissue electron density information required by radiation dose calculation, proposed in the 2000s, 4D-CT has become the standard imaging technique of treatment moving target [7]. **Figure 2** shows a diagram of 4D-CT. For a simple description, 4D-CT samples projections repetitively at a couch position for at least one respiratory cycle before moving to the next imaging position. With synchronized respiratory cycle information, all projections are retrospectively sorted into

**Figure 2.**  *A simple diagram of 4D-CT implementation.* 

#### *4D-MRI in Radiotherapy DOI: http://dx.doi.org/10.5772/intechopen.84592*

 different phase bins (typically 10) that consists of an averaged respiratory cycle derived from the whole scan period [8]. The respiratory cycle information is usually derived by recording 2D trajectories of external surrogates, such as inferred reflection markers and pneumatic bellows [9, 10]. As a result, multiple 3D CT volumes from different phase volumes are reconstructed. By delineating CTV volumes at each 3D volume, ITV can be generated as a union of CTVs from all phases with a possible small margin for positioning uncertainty.

 Despite of its popularity, 4D-CT has a few problems: (1) when patient's breath becomes irregular in terms of amplitude and period length, abrupt changes in projections may lead to motion-induced artifacts in the reconstructed phase volumes. **Figure 3** shows a reconstructed phase volume in 4D-CT. The diaphragm boundary discontinuity with a ghost displacement indicated by red arrow is a typical rendering of motion artifacts in 4D-CT; (2) although trajectories of external surrogates and internal organs are correlated [11], potential hysteresis between the two trajectories may impact overall treatment accuracy [12]; and (3) 4D-CT requires extended scan time and leads to increased imaging radiation dose as a potential patient health risk.

In addition to 4D-CT, ultrasound has been utilized for motion assessment in radiotherapy because of its fast imaging time and relatively simple implementation. Earlier application of ultrasound focused at localizing displaced prostate radiotherapy volume through transrectal imaging [13]. To image respiratory motion, previous work reported its use for upper abdominal radiotherapy. In addition to assess motion at simulation, same imaging strategy can be deployed at each treatment fraction for real-time verification [14, 15]. However, motion assessment using ultrasound sometime can only be achieved by indirect measurement of a nearby landmark instead of target volume [16]. The generally poor visualization capability of ultrasound limits its utilization in the current practice of radiotherapy.

Following the increased utilization of MRI in radiotherapy, 4D-MRI has become a popular area in image-guided radiotherapy (IGRT) after the birth of 4D-CT. Generally, MRI has excellent soft-tissue contrast with zero radiation hazard in comparison to 4D-CT. 4D-MRI is thus highly desirable in the radiotherapy workflow. In the last decade, many works have been done for the development of 4D-MRI. At present, however, there is no fully established 4D-MRI technique by major vendors in radiotherapy clinic. Implementation of 4D-MRI in clinic is still at investigational stage due to excessive technical involvement. Nevertheless, current results of 4D-MRI application showed its promising value in the era of IGRT [17].

In the following section, we will introduce 4D-MRI basic theories and current technologies and discuss emerging topics in 4D-MRI research and development. Radiotherapy application of 4D-MRI will be discussed based on our clinical experience in another section.

**Figure 3.**  *An example of motion artifacts in 4D-CT.* 

#### **2. 4D-MRI: basic theories and technologies**

In this section, we first describe the brief history of 4D-MRI debut and then discuss basic theories and implementation technologies of current 4D-MRI methods. Frontier research topics in 4D-MRI are introduced in the rear part in this section.

#### **2.1 Fast imaging: early effort**

Earliest effort for MRI-based motion quantification focused on 2D-based cine imaging, i.e., continuously acquire images at a fixed 2D coordinate frame. The acquired 2D images could be stacked as a movie that described target motion. This requires fast imaging sequences to achieve high frame rates. Several approaches were employed. Koch et al. first used gradient echo technique in lung motion imaging [18]. Shimizu et al. imaged liver tumor motion with multiple T1-weighted gradient slices [19]. Kirilova et al. used T2-weighted single shot fast spin echo sequence for liver tumor position tracking [20]. A handful of works adopted balanced steady-state free precession (bSSFP) imaging on three orthogonal planes for tumor as well as diaphragm motion tracking [21, 22]. The reported frame rate using bSSFP could be up to 10 frames per second.

 The obvious drawback of 2D cine imaging is the deficiency of volumetric motion capture, which is critical when motion pattern is heterogeneous within the imaging region of interest (ROI). Repetitive 3D volume acquisition (i.e., real-time 4D-MRI) was then investigated since it yielded truly real-time volumetric imaging without additional post-processing. This approach is usually accomplished with parallel imaging with a trade of image quality. Blackall et al. used fast file echo with echo planar imaging (FFE-EPI) for real-time 4D-MRI implementation for lung RT planning [23]. In spite of its reported high temporal resolution (up to 330 ms/frame), the acquired image showed considerably less vessel structure details within the lung. Dinkel et al. implemented TREAT sequence, a 3D time-resolved echo shared gradient echo sequence with parallel imaging [24]. This technique achieved a frame rate of 1400 ms/frame covering a large field of view (FOV) (400 × 400 × 300 mm) with a relatively low spatial resolution (voxel size 3.1 × 3.1 × 4 mm). Tryggestad et al. reported their real-time 4D-MRI protocol with frame rate 1000 ms/frame at the cost of lower signal-to-noise ratio (SNR) and intrascan motion [25]. However, these works did not achieve high temporal resolution and acceptable spatial resolution at the same time. Since a typical human's breathing cycle period is about 3–5 seconds, high temporal resolution <500 ms/frame is desired for multiphase reconstruction. In addition, high spatial resolution at the level of 1 mm isotropic voxel size might be necessary for small size SBRT target in lung/liver treatment, which could be 1 cc or less. Currently available MR scanner capabilities limit further improvement of real-time 4D-MRI.

#### **2.2 Retrospective sorting: 2D-based**

 Retrospective sorting has been the mainstream technique for 4D-MRI reconstruction. Similar to the 4D-CT implementation, retrospective sorting in 4D-MRI records a motion surrogate's trajectory during the scan. Acquisition data are sorted into different bins based on the amplitude/phase information of the surrogate trajectory. To date, retrospective sorting with multislice dynamic 2D acquisition (2D-based sorting) is commonly reported for clinical and preclinical investigations. In short, 2D MRI image slices at interleaved slice locations were sorted accordingly before 3D volume stacking. By theory, this strategy can achieve high spatial resolution while maintaining high in-plane spatial resolution. On the other hand, retrospective sorting requires intensive image post-processing offline with high software demand.

#### *4D-MRI in Radiotherapy DOI: http://dx.doi.org/10.5772/intechopen.84592*

So far, several works of 2D-based sorting techniques have been reported with different motion surrogate selection and image acquisition sequence. von Siebenthal et al. invented a navigator-based sorting method. In this method, navigator slices were acquired at a fixed location, interleaved with slice locations that were sequentially stepped through the imaging volume. 2D images were acquired using 2D bSSFP sequence repetitively for nearly 1 hour. Image sorting was carried based on navigator slice similarity using a cost function that combines directional shifts of image registration [26]. This sophisticated scheme achieve 180 ms/frame temporal resolution and 1.8 × 1.8 mm in-plane resolution, but the navigator acquisition prolonged total scan time and did not guarantee measurement reproducibility.

Remmert et al. investigated the feasibility of using respiratory surrogate. A rapid imaging sequence using 2D fast low-angle shot (FLASH) with generalized auto-calibrating partially parallel acquisition (GRAPPA) was adopted on Cartesian grid. Respiratory surrogate was extracted as the positioning of the piston rod of the imaged dynamic porcine lung phantom [27]. The results demonstrated the feasibility of respiratory surrogate sorting. Hu et al. implemented a respiratory-triggered 4D-MRI reconstruction. The respiratory amplitude was derived from a turbo spin echo sequence, and 2D image acquisition was achieved by T2-weighted EPI [28]. The in vivo demonstration of this technique in healthy volunteers successfully reconstructed four phase volumes.

Other surrogates are also proposed for convenient sorting implementation. Cai et al. proposed a sorting surrogate based on body area (BA), which was defined as the area inside the binary mask of body contour from 2D fast steady-state acquisition (FIESTA). The surrogate trajectory was determined by excluding the low-frequency change of BA as subject anatomic change [29]. Tryggestad et al. integrated signals from Physiologic Monitoring Unit (PMU) by Siemens for retrospective sorting. This PMU signal was derived from a pneumatic device attached to the subject's upper abdomen. Acquired at 50 Hz sampling rate, the PMU signal was synchronized with 2D repetitive acquisition [25].

In spite of its popularity, 2D-based sorting methods suffer from two major problems: (1) large slice thickness (typically 5–10 mm) may not be sufficient to quantify target motion when the target is small, and the reconstructed volume may not look like continuous on slice direction; and (2) the reconstructed images tend to have stitch motion artifacts similar to those in 4D-CT. **Figure 4** shows an example of stitch motion artifacts. This resorting artifact of radiant rays is caused by unpredicted motion change during image acquisition.

#### **2.3 Retrospective sorting: 3D-based**

3D-based retrospective sorting has gained its attention with advances of both imaging hardware and software technologies. Similar to 2D-based equivalent, 3D-based retrospective sorting requires motion surrogate recording synchronized with data acquisition. The most distinguished feature of 3D-based sorting is that raw k-space acquisitions, rather than 2D image data in real space, are sorted into different phase volume bins. Image reconstruction in 3D fashion (but could also be done in 2D) has to be performed after retrospective sorting. Image quality can be improved with advanced image reconstruction technique with possible isotropic voxel size. Generally, 3D-based retrospective sorting requires more sophisticated data processing algorithms and more demanding hardware manipulation. Nevertheless, 3D-based retrospective sorting has become the selection by most recent 4D-MRI developments.

In 2008, Tokuda et al. proposed an early trial of 3D-based retrospective sorting. The acquired k-space echo was filled into multiple assigned bins based on

#### **Figure 4.**

*An example of stitch motion artifacts in reconstructed 4D-MRI. This image was reconstructed from computer simulation instead of human subject imaging.* 

amplitude ranges of respiratory surrogate, which was implemented by a navigator echo acquired at a configurable sampling rate [30]. In a patient study, 4D volumes were successfully reconstructed using 10.5 min acquisition time with compromised image quality compared with breath-hold (BH) 3D acquisitions.

 In the last several years, several representative works with 3D-based sorting have been reported as most recent advances in 4D-MRI. Deng et al. utilized a continuous spoiled gradient echo sequence with 3D radial trajectory for fast volumetric acquisition [31]. Respiratory surrogate was derived as "self-gating" (SG) measurement, which was 1D Fourier transform of SG acquisition lines through the k-space center (kz direction). Acquisitions were retrospectively sorted into ten phase volumes using phase percentage information on surrogate trajectory, and image reconstruction was carried out with a conjugate-gradient sensitivity encoding (CG-SENSE) with self-sensitivity calibration [32].

Similarly, Feng et al. proposed a 4D-MRI framework XD-GRASP (extra-dimension golden-angle radial sparse parallel) technique, which included 3D radial-based acquisition trajectory and surrogate with SG [33]. Principal component analysis (PCA) could be used on SG surrogate to separate cardiac motion or MR contrast uptake from respiratory motion. Each 3D volume was iteratively reconstructed on radial grid.

 Acquisition on Cartesian grid has also been reported. Han et al. invented a rotating 3D Cartesian k-space recording method (ROCK) for 4D-MRI acquisition [34]. This method simulated a quasispiral acquisition trajectory with varying sampling densities on radial direction. Balanced SSFP sequence was chosen for its improved SNR over gradient recalled echo technique, and amplitude-based sorting was based on SG motion surrogate. Wang et al. also implemented a sparse Cartesian-based acquisition trajectory simulating a multi-ray profile regulated by golden-ratio angular increment [35].

#### **2.4 Discussion: technical factors in 4D-MRI**

As a summary of current 4D-MRI works (with a focus on 3D-based retrospective sorting approach), we hereby discuss a few key technical factors that one may need to consider for developing a 4D-MRI method:

#### *2.4.1 Acquisition protocol*

K-space acquisition can be realized by different techniques since retrospective sorting does not directly depend on data acquisition scheme. Both radial-based and Cartesian-based trajectories are valid for acquisition. Since radial-based trajectories have higher sampling density near k-space center (low-frequency component) in their nature, they might be preferred for fast 3D acquisition. Golden-ratio means or its equivalent technologies are commonly used for 4D-MRI since the azimuthal increments are relatively constant after volume sorting [36–38]. However, radial-based trajectories may lead to severe motion artifacts after retrospective sorting when the subject's motion is irregular [34]. Cartesian-based trajectories could simulate k-space central/peripheral sampling weights in radial-trajectories and might be easier for image reconstruction without data regridding. Nevertheless, both approaches require extensive hardware/software editing that most clinical MR units may not be fully ready for.

To implement SG, low-frequency component in k-space has to be sampled repeatedly during the total acquisition time. As discussed above, SG can be derived as 1D projection on kz direction through k-space center. SG signal can capture displacement on superior-inferior (SI) direction that is sensitive to respiratory motion [39]. SG can also be derived as phase shift measurement of 0-frequency (DC) point at k-space center [35]. It has to be pointed out that k-space center does not have to be sampled periodically as long as the sampling intervals are known. Since the human breath is mostly modeled as sinusoid waveform, it is more straightforward to sample k-space center at certain rhythms.

#### *2.4.2 Retrospective sorting practice*

In retrospective sorting, data binning can use either surrogate's amplitude or phase percentage (temporal location within a breath cycle) information. Both approaches are valid in both 4D-MRI and 4D-CT. However, when the subject's breathing is not regular, potential error could be made in retrospective sorting [40, 41]. In clinical practice, intrascan variation of both breathing amplitude and period is commonly seen. There is no established theory regarding variation theory, but breathing period is subject to more change when the scan time is long for pulmonary function compromised subjects. Drastic change of surrogate amplitude, however, is usually caused by random event such as cough.

To reduce irregularity-induced data divergence, a straightforward way is to employ external motion management devices as in 4D-CT imaging, such as abdominal compression and body vacuum bag. However, such devices can be cumbersome for MR imaging protocols.

During retrospective sorting, one determines certain threshold values and excludes data acquisitions when surrogate trajectory is out of range [29]. In spite of its simplicity, this approach may reduce data utility when the subject's breathing is irregular, which may further lead to undersampling artifacts in the reconstructed images. A soft-gating approach has been reported for amplitude-based on a Gaussian weighting function with its Full width at half maximum (FWHM) determined as a function of surrogate motion range [34]. This approach can improve data utilization with improved SNR in the reconstructed images [42, 43].

 Wang et al. designed a spatiotemporal index (STI) as a quadratic sum of amplitude and phase discrepancies in retrospective sorting [35]. Each acquisition can be used for reconstructing multiple phase volumes when discrepancy criteria are met. In combination with the selected acquisition trajectory, such criteria were designed as tight rules near k-space center and loose rules near k-space periphery to further improve data utilization.

#### *2.4.3 Image reconstruction*

Image reconstruction has become a focused topic in 4D-MRI because of sparse k-space data after 3D-based retrospective sorting. With the advent of compressed sensing, i.e., extract compressible signal from undersampled data, many iterative MR reconstruction algorithms have been successfully developed [44–46]. In the specific 4D-MRI reconstruction, L1-norm (total variation [47]), L2-norm (total generalized variation, TGV [48]; Tikhonov regularization [49]), and wavelet (Daubechies [50]) regularization algorithms have been utilized. Detailed mathematics of these algorithms go beyond the scope of this discussion.

A unique feature of 4D-MRI reconstruction is the potential implementation of spatiotemporal constrained reconstruction. Because of averaging nature in retrospective sorting, motion continuity is usually guaranteed after reconstructing each phase volume as an independent 3D volume. A possible improvement is view sharing, which enable the use of same data in multiple volumes. Such technique has been widely used in dynamic MRI imaging for pharmacokinetics study [51, 52]. Because of motion sensitivity of low-frequency k-space component, view sharing of high-frequency component in combination with iterative reconstruction is a viable solution [35]. Temporal constraint can also be explicitly written as penalty terms in reconstruction. Total variation representation of finite motion differences has been adopted [33, 34].

#### **2.5 Emerging topics in 4D-MRI research**

 Trends in 4D-MRI research have been following evolving technologies in radiation oncology. The most relative topic is to use deformable image registration (DIR) to model anatomic motion during respiratory cycles. A detailed description of DIR cannot be included in this section; for a short discussion, DIR tries to wrap one image to another one with much more transform degrees that are different across the ROI instead of 6 rigid transform degrees. A registration is represented by a deformation vector field (DVF) in the same size as the source image, which points each single voxel in source image to its destination in the coordinate system of target image. DIR has been used in radiotherapy for time-series image registration, image outcome for treatment assessment, and dose wrapping for adaptive therapy [53].

In 4D-MRI, the motion-induced anatomic change can be seen as a deformation process. Each phase volume (2D or 3D) is a deformation from a standard reference volume. This reference volume can be a stable phase volume in 4D series (such as endof-exhalation (EOE) phase) or a breath-hold (BH) volume. Thus, respiratory motion can be described by a series of DVFs derived from DIR. This enables the motion information transfer to recreate respiratory motion derived from one MR contrast to another one [54].

 Recently, Harris et al. and Stemkens et al. proposed their similar approaches of 4D-MRI reconstruction based on DIR manipulation and a priori patient-specific motion model [55, 56]. When a 4D-MRI series for a patient is available, a series of DVFs could be generated by DIR. Principal component analysis (PCA) was used to decompose this DVF series into three principle components (PCs). Any future motion imaging could be seen as a simple weighted sum of these three PCs. To derive the three weighting coefficients, the acquisition could be done rapidly by single or multislice 2D acquisition, and the coefficients were solved as a minimization problem based on the similarity of the reconstructed 4D-MRI series at fixed position(s) with 2D acquisition(s) [57]. Because of rapid 2D acquisition, this method could generate volumetric images in cine fashion up to 20 frames/s [57]. Harris et al. also reported that such rapid 2D acquisition can be done by planar KV fluoro images which are widely available on modern radiation linear accelerators

#### *4D-MRI in Radiotherapy DOI: http://dx.doi.org/10.5772/intechopen.84592*

(LINAC) [58, 59]. This point attracts the attention since MR-based radiotherapy treatment guidance can be realized on current radiotherapy platform.

It is crucial to ensure the DIR accuracy when generating the DVF. Li et al. reported their time-resolved 4D-MRI (TR-4DMRI) method with improved DIR reconstruction. Based on the a priori motion information, a pseudo demon force was introduced and applied to the coarse volumetric alignment. The fine-tuning of DIR was performed at multiple resolutions by demon forces [60]. This method was argued with better handle of large anatomy deformation during possible irregular breath. Digital phantom results showed that this technique successfully reconstructed fast 4D series and identified some questionable cases without missing true negative.

The wave of artificial intelligence (AI) in medicine (specifically in radiation oncology imaging) has enlightened many sparkling ideas in medical imaging research [61]. Together with the use of AI in pattern recognition from large-scale data, the currently required patient-specific motion pattern can be potentially derived from population results. Thus, 4D-MRI based on DIR modeling may become the focus of 4D-MRI research in the next decade.

#### **3. 4D-MRI: clinical application**

Although 4D-MRI has been demonstrated with great potential of anatomic motion quantification, its application in clinic is still premature for standard practice. Nevertheless, a few works have reported 4D-MRI's values in radiotherapy clinic. In this section, we discuss possible clinical applications of 4D-MRI in radiotherapy.

#### **3.1 Radiotherapy target delineation**

Similar to the 4D-CT application, 4D-MRI provides motion direction and quantified motion range information for ITV generation at the same level of accuracy as 4D-CT [62]. Because of superior soft tissue contrast, 4D-MRI may better illustrate ITV definition with finer anatomical detail boundaries.

 **Figure 5** shows an example of 4D-MRI-based ITV delineation for liver SBRT in our clinic. This case was about liver mets from breast cancer, and the prescription was 16Gy × 3 fractions. 4D-MRI was acquired by bSSFP sequence with 2D-based retrospective sorting [29]. Interslice distance was 5 mm. A total of 10 phase volumes (0, 10, … 90%) were reconstructed. GTV was contoured in each phase volume as shown by red contours in five representative phase volumes in **Figure 5**. The reference frame of 4D-MRI was registered to the planning CT volume by rigid transform. ITV-MRI, the union of 10 GTVs from 4D-MRI, was mapped to the planning CT as illustrated. The final ITV for this case was adjusted by our radiation oncologist to combine information from both MRI and CT studies.

#### **3.2 Onboard treatment guidance**

Since 2010, MR-guided radiotherapy units have become commercially available. The integrated onboard MR imaging capability can provide potentially improved patient positioning accuracy with affluent soft tissue details. In addition, intratreatment imaging during radiation enables image-based treatment gating, which could reduce margin size for PTV definition with potentially reduced normal tissue toxicity [63]. Currently, both low-field-strength (0.35 T) unit (ViewRay, Oakwood Village, Ohio) and high-field-strength (1.5 T) unit (Elekta AB, Stockholm, Sweden) have started treating patient.

**Figure 5.**  *A clinical example of 4D-MRI in liver SBRT target delineation.* 

#### **Figure 6.**

*VC-MRI for onboard MR imaging accuracy simulation. MRIprior, reference EOE volume in pre-treatment 4D-MRI simulation; VCMRIgt, ground truth EOI volume at treatment day for onboard match; and VCMRIest, estimated EOI volume from the proposed VC-MRI approach.* 

#### *4D-MRI in Radiotherapy DOI: http://dx.doi.org/10.5772/intechopen.84592*

Onboard image guidance on MR-guided radiotherapy units are implemented by 2D imaging on orthogonal views. Volumetric imaging, though technically feasible, are not realistic because of long imaging time. At this moment, no in vivo implementation of onboard 4D-MRI guidance has been reported. Nevertheless, a few works have tried to demonstrate its feasibility. Harris et al. used digital XCAT phantom simulation to examine their VC-MRI technique for onboard patient positioning accuracy [56]. As shown in **Figure 6**, the estimated VC-MRI volume at end-of-inhalation (EOI) phase (third column) was accurate when comparing to the ground truth volume (second column). The reported target center-of-mass-shift (COMS) was about 1 mm or less on SI direction in most simulation scenarios.

Han et al. implemented their ROCK 4D-MRI technique on their 0.35 T-MRguided radiotherapy unit [64]. Seven patients with abdominal tumors were imaged with both ROCK 4D-MRI and 2D-cine (reference) techniques. Because of relatively long imaging time (~10 min) in ROCK, image acquisition was performed after treatment as feasibility studies. They reported that when compared with reference 2D-cine results, motion quantification in 4D-MRI was about 1 mm different on SI direction. 3D anatomical details were successfully rendered without motion artifacts for onboard imaging. Optimistically speaking, 4D-MRI for pre-treatment patient positioning could become available in the next decade with novel 4D-MRI methods and improved hardware developments.

### **Author details**

Chunhao Wang\* and Fang-Fang Yin Department of Radiation Oncology, Duke University Medical Center, Durham, NC, United States

\*Address all correspondence to: chunhao.wang@duke.edu

© 2019 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/ by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

### **References**

[1] Khan FM, Gibbons JP. Khan's the Physics of Radiation Therapy. Baltimore, MD: Lippincott Williams & Wilkins; 2014

[2] Landberg T, Chauaudra J, Dobbs H. Prescribing, Recording and Reporting Photon Beam Therapy: ICRU Report 50. Maryland, USA: International Commission on Radiotherapy Units and Measurements; 1993

[3] Ozhasoglu C, Murphy MJ. Issues in respiratory motion compensation during external-beam radiotherapy. International Journal of Radiation Oncology, Biology, Physics. 2002;**52**(5):1389-1399

[4] Stroom JC, Heijmen BJ. Geometrical uncertainties, radiotherapy planning margins, and the ICRU-62 report. Radiotherapy and Oncology. 2002;**64**(1):75-83

[5] Jin J-Y et al. A technique of using gated-CT images to determine internal target volume (ITV) for fractionated stereotactic lung radiotherapy. Radiotherapy and Oncology. 2006;**78**(2):177-184

[6] Siewerdsen J et al. Multimode C-arm fluoroscopy, tomosynthesis, and cone-beam CT for image-guided interventions: From proof of principle to patient protocols. In: Medical Imaging 2007: Physics of Medical Imaging. San Diego, CA: International Society for Optics and Photonics; 2007

[7] Low DA et al. A method for the reconstruction of four-dimensional synchronized CT scans acquired during free breathing. Medical Physics. 2003;**30**(6):1254-1263

[8] Pan T et al. 4D-CT imaging of a volume influenced by respiratory

motion on multi-slice CT. Medical Physics. 2004;**31**(2):333-340

[9] Glide-Hurst CK et al. Evaluation of two synchronized external surrogates for 4D CT sorting. Journal of Applied Clinical Medical Physics. 2013;**14**(6):117-132

 [10] Quirk S, Becker N, Smith W. External respiratory motion analysis and statistics for patients and volunteers. Journal of Applied Clinical Medical Physics. 2013;**14**(2):90-101

[11] Hoisak JD et al. Correlation of lung tumor motion with external surrogate indicators of respiration. International Journal of Radiation Oncology, Biology, Physics. 2004;**60**(4):1298-1306

[12] Ruan D et al. Inference of hysteretic respiratory tumor motion from external surrogates: A state augmentation approach. Physics in Medicine & Biology. 2008;**53**(11):2923

[13] Lattanzi J et al. Ultrasound-based stereotactic guidance of precision conformal external beam radiation therapy in clinically localized prostate cancer. Urology. 2000;**55**(1):73-78

[14] Fuss M et al. Daily ultrasoundbased image-guided targeting for radiotherapy of upper abdominal malignancies. International Journal of Radiation Oncology, Biology, Physics. 2004;**59**(4):1245-1256

[15] O'shea T, Bamber J, Harris E. MO-DE-210-05: Improved accuracy of liver feature motion estimation in B-Mode ultrasound for image-guided radiation therapy. Medical Physics. 2015;**42**(6Part28):3560-3560

[16] Mason SA et al. Towards ultrasound-guided adaptive radiotherapy for cervical cancer:

#### *4D-MRI in Radiotherapy DOI: http://dx.doi.org/10.5772/intechopen.84592*

Evaluation of Elekta's semiautomated uterine segmentation method on 3D ultrasound images. Medical Physics. 2017;**44**(7):3630-3638

[17] Dawson LA, Sharpe MB. Imageguided radiotherapy: Rationale, benefits, and limitations. The Lancet Oncology. 2006;**7**(10):848-858

 [18] Koch N et al. Evaluation of internal lung motion for respiratory-gated radiotherapy using MRI: Part I correlating internal lung motion with skin fiducial motion. International Journal of Radiation Oncology, Biology, Physics. 2004;**60**(5):1459-1472

[19] Shimizu S et al. High-speed magnetic resonance imaging for fourdimensional treatment planning of conformal radiotherapy of moving body tumors1. International Journal of Radiation Oncology, Biology, Physics. 2000;**48**(2):471-474

[20] Kirilova A et al. Threedimensional motion of liver tumors using cine-magnetic resonance imaging. International Journal of Radiation Oncology, Biology, Physics. 2008;**71**(4):1189-1195

[21] Feng M et al. Characterization of pancreatic tumor motion using cine MRI: Surrogates for tumor position should be used with caution. International Journal of Radiation Oncology, Biology, Physics. 2009;**74**(3):884-891

[22] Cai J et al. Evaluation of the reproducibility of lung motion probability distribution function (PDF) using dynamic MRI. Physics in Medicine & Biology. 2006;**52**(2):365

[23] Blackall J et al. MRI-based measurements of respiratory motion variability and assessment of imaging strategies for radiotherapy planning. Physics in Medicine & Biology. 2006;**51**(17):4147

[24] Dinkel J et al. 4D-MRI analysis of lung tumor motion in patients with hemidiaphragmatic paralysis. Radiotherapy and Oncology. 2009;**91**(3):449-454

[25] Tryggestad E et al. Respirationbased sorting of dynamic MRI to derive representative 4D-MRI for radiotherapy planning. Medical Physics. 2013;**40**(5):051909

[26] von Siebenthal M et al. 4D MR imaging of respiratory organ motion and its variability. Physics in Medicine & Biology. 2007;**52**(6):1547

[27] Remmert G et al. Four-dimensional magnetic resonance imaging for the determination of tumour movement and its evaluation using a dynamic porcine lung phantom. Physics in Medicine & Biology. 2007;**52**(18):N401

[28] Hu Y et al. Respiratory amplitude guided 4-dimensional magnetic resonance imaging. International Journal of Radiation Oncology Biology Physics. 2013;**86**(1):198-204

[29] Cai J et al. Four-dimensional magnetic resonance imaging (4D-MRI) using image-based respiratory surrogate: A feasibility study. Medical Physics. 2011;**38**(12):6384-6394

[30] Tokuda J et al. Adaptive 4D MR imaging using navigator-based respiratory signal for MRI-guided therapy. Magnetic Resonance in Medicine. 2008;**59**(5):1051-1061

[31] Deng Z et al. Four-dimensional MRI using three-dimensional radial sampling with respiratory self-gating to characterize temporal phase-resolved respiratory motion in the abdomen. Magnetic Resonance in Medicine. 2016;**75**(4):1574-1585

[32] Pang J et al. Accelerated whole-heart coronary MRA using motion-corrected

sensitivity encoding with threedimensional projection reconstruction. Magnetic Resonance in Medicine. 2015;**73**(1):284-291

 [33] Feng L et al. XD-GRASP: Goldenangle radial MRI with reconstruction of extra motion-state dimensions using compressed sensing. Magnetic Resonance in Medicine. 2016;**75**(2):775-788

[34] Han F et al. Respiratory motionresolved, self-gated 4D-MRI using rotating cartesian k-space (ROCK). Medical Physics. 2017;**44**(4):1359-1368

[35] Wang C et al. A spatiotemporalconstrained sorting method for motion-robust 4D-MRI: A feasibility study. International Journal of Radiation Oncology, Biology, Physics. 2018;**103**(3):758-766

[36] Kazantsev IG, Matej S, Lewitt RM. Optimal ordering of projections using permutation matrices and angles between projection subspaces. Electronic Notes in Discrete Mathematics. 2005;**20**:205-216

[37] Tsai CM, Nishimura DG. Reduced aliasing artifacts using variabledensity k-space sampling trajectories. Magnetic Resonance in Medicine. 2000;**43**(3):452-458

[38] Chan RW et al. Temporal stability of adaptive 3D radial MRI using multidimensional golden means. Magnetic Resonance in Medicine. 2009;**61**(2):354-363

[39] Pang J et al. ECG and navigator-free four-dimensional whole-heart coronary MRA for simultaneous visualization of cardiac anatomy and function. Magnetic Resonance in Medicine. 2014;**72**(5):1208-1217

[40] Jiang SB. Technical aspects of image-guided respiration-gated radiation therapy. Medical Dosimetry. 2006;**31**(2):141-151

[41] Lu W et al. A comparison between amplitude sorting and phase-angle sorting using external respiratory measurement for 4D CT. Medical Physics. 2006;**33**(8):2964-2974

[42] Cheng JY et al. Free-breathing pediatric MRI with nonrigid motion correction and acceleration. Journal of Magnetic Resonance Imaging. 2015;**42**(2):407-420

[43] Uecker M et al. ESPIRiT—an eigenvalue approach to autocalibrating parallel MRI: Where SENSE meets GRAPPA. Magnetic Resonance in Medicine. 2014;**71**(3):990-1001

[44] Gamper U, Boesiger P, Kozerke S. Compressed sensing in dynamic MRI. Magnetic Resonance in Medicine. 2008;**59**(2):365-373

[45] Lustig M, Donoho D, Pauly JM. Sparse MRI: The application of compressed sensing for rapid MR imaging. Magnetic Resonance in Medicine. 2007;**58**(6):1182-1195

[46] Donoho DL. Compressed sensing. IEEE Transactions on Information Theory. 2006;**52**(4):1289-1306

[47] Block KT, Uecker M, Frahm J. Undersampled radial MRI with multiple coils. Iterative image reconstruction using a total variation constraint. Magnetic Resonance in Medicine. 2007;**57**(6):1086-1098

[48] Knoll F et al. Second order total generalized variation (TGV) for MRI. Magnetic Resonance in Medicine. 2011;**65**(2):480-491

[49] Ying L, Xu D, Liang Z-P. On Tikhonov regularization for image reconstruction in parallel MRI. In: Engineering in Medicine and Biology Society, 2004. IEMBS'04. 26th Annual International Conference of the IEEE. IEEE; 2004

#### *4D-MRI in Radiotherapy DOI: http://dx.doi.org/10.5772/intechopen.84592*

 [50] Daubechies I, Defrise M, De Mol C. An iterative thresholding algorithm for linear inverse problems with a sparsity constraint. Communications on Pure and Applied Mathematics. 2004;**57**(11):1413-1457

[51] Wang C et al. Accelerated brain DCE-MRI using iterative reconstruction with total generalized variation penalty for quantitative pharmacokinetic analysis: A feasibility study. Technology in Cancer Research & Treatment. 2017;**16**(4):446-460

[52] Subashi E et al. A comparison of radial keyhole strategies for high spatial and temporal resolution 4D contrast-enhanced MRI in small animal tumor models. Medical Physics. 2013;**40**(2):022304

[53] Brock KK, Consortium DRA. Results of a multi-institution deformable registration accuracy study (MIDRAS). International Journal of Radiation Oncology, Biology, Physics. 2010;**76**(2):583-596

[54] Yang J et al. Four-dimensional magnetic resonance imaging using axial body area as respiratory surrogate: Initial patient results. International Journal of Radiation Oncology, Biology, Physics. 2014;**88**(4):907-912

[55] Stemkens B et al. Image-driven, model-based 3D abdominal motion estimation for MR-guided radiotherapy. Physics in Medicine & Biology. 2016;**61**(14):5335

[56] Harris W et al. A technique for generating volumetric cine-magnetic resonance imaging. International Journal of Radiation Oncology, Biology, Physics. 2016;**95**(2):844-853

[57] Wang C et al. TU-AB-BRA-09: A novel method of generating ultrafast volumetric cine MRI (VC-MRI) using prior 4D-MRI and on-board phase-skipped encoding

acquisition for radiotherapy target localization. Medical Physics. 2016;**43**(6Part33):3735-3735

[58] Wang C et al. A novel method of generating onboard 4D-MRI for liver SBRT target localization using prior 4D-MRI simulation and onboard limited angle kV acquisition from a conventional LINAC. International Journal of Radiation Oncology Biology Physics. 2017;**99**(2):S126-S127

[59] Harris W et al. A novel method to generate on-board 4D MRI using prior 4D MRI and on-board kV projections from a conventional LINAC for target localization in liver SBRT. Medical Physics. 2018;**45**(7):3238-3245

[60] Li G et al. Introduction of a pseudo demons force to enhance deformation range for robust reconstruction of super-resolution time-resolved 4 DMRI. Medical Physics. 2018;**45**(11):5197-5207

[61] Thompson RF et al. Artificial intelligence in radiation oncology imaging. International Journal of Radiation Oncology Biology Physics. 2018;**102**(4):1159-1161

[62] Oar A et al. Comparison of four dimensional computed tomography and magnetic resonance imaging in abdominal radiotherapy planning. Physics and Imaging in Radiation Oncology. 2018;**7**:70-75

[63] Lagendijk JJ, Raaymakers BW, Van Vulpen M. The magnetic resonance imaging–linac system. Seminars in Radiation Oncology. Philadelphia, PA: WB Saunders; 2014;**24**(3)

[64] Han F et al. Respiratory motionresolved, self-gated 4D-MRI using rotating Cartesian K-space (ROCK): Initial clinical experience on an MRI-guided radiotherapy system. Radiotherapy and Oncology. 2018;**127**(3):467-473

## *Edited by Lachezar Manchev*

Diagnostic imaging has undergone many changes over the last several years. Technical developments have defned Magnetic Resonance Imaging (MRI) as the leading diagnostic modality in diferent diseases. MRI is defnitive and sensitive and the current requirements of medicine call for radiologists to be profcient in its use. Tis book provides complete and detailed information about the fast-developing feld of MRI from physicians, radiologists, and other clinical specialists. It is a practical guide to using MRI in areas such as cardiology and pulmonology, among others.

Published in London, UK © 2019 IntechOpen © semnic / iStock