Keywords

1 Introduction

The mechanical property of a tissue is related to physiological and pathological states. Magnetic resonance elastography (MRE) and ultrasound elastography (USE) are imaging techniques that non-invasively quantify the mechanical properties of tissue by using magnetic resonance imaging and ultrasound imaging systems [1, 2]. It is expected that measuring the mechanical properties of tissues will be useful in the diagnosis of diseases such as hepatic fibrosis and cancer. MRE visualizes shear-wave patterns within a tissue using a modified phase-contrast MR sequence. In order to generate shear waves within the tissue, external vibration systems are used (Fig. 39.1). The local quantitative values of tissue viscoelasticity (stiffness) are calculated from the shear-wave pattern by using an inversion algorithm [3,4,5]. We have been developing an MRE system using clinical MRI and MR microscope for measuring viscoelasticity at multi-scale and multi-frequency (Fig. 39.2) [6, 7].

Fig. 39.1
figure 1

Schematic of the MR elastography system

Full size image
Fig. 39.2
figure 2

Multi-modality equipment for measuring viscoelasticity at multi-scale and multi-frequency

Full size image

The measured viscoelasticity can be used as an imaging biomarker [8]. A quantitative phantom is required to assess the accuracy and repeatability of elastography systems. We have been developing tough and stable polyacrylamide (PAAm) gel phantoms for this purpose [9,10,11,12]. We developed a phantom with viscoelasticity close to that of living tissue by using glycerin as a solvent. For ultrasonic measurements, a scatterer is necessary for the phantom. The material and concentration of the scatterer are related to the stability of ultrasonic measurement and uniformity of the MRI image.

In this study, we compared the mechanical properties of our phantoms with those of living tissue and optimized the scatterer material and concentration for ultrasound measurement with the developed MRE system and commercial USE system.

1.1 Development of a Tissue-Mimicking Viscoelastic Phantom for Quantitative Assessment of MRE

We fabricated gel phantom sheets with high loss moduli (G″). The phantom sheets were designed to have a storage modulus (G′) of 1.5 kPa and a loss tangent (tan δ = G″/G′) greater than 0.2 at 60 Hz [13]. The storage and loss moduli of the phantom and a fresh in vitro bovine liver were measured using a parallel-disc rheometer (MCR302, Anton-Parr). We repeated the measurements on these three samples with the strain amplitude set to 1% (Table 39.1).

Table 39.1 Rheometer measurement parameters
Full size table

The phantom (diameter: 120 mm, height: 150 mm) was measured with MRE using 3-T MRI (MAGNETOM Skyra, Siemens) at 30, 40, 50, and 60 Hz (Table 39.2) by spin-echo-EPI-MRE pulse sequence (work in progress) and the vibration from a pneumatic driver system with active driver to a passive driver to create the shear wave in the phantom. The passive driver with a diameter of 18 cm was placed on the center of the phantom. The fresh in vitro bovine liver (width: 135 mm, height: 70 mm, length: 250 mm) was measured with MRE using 0.3-T MRI (Hitachi) at 31.25, 62.5, 100, and 200 Hz (Table 39.3) by spin-echo-EPI-MRE pulse sequence (a motion encoding gradient (MEG) was added to SE-EPI using the sequence development environment of Hitachi) and ultrasound system (ACUSON S3000, Siemens) with virtual touch IQ (VTIQ). At high-frequency measurement of the fresh in vitro bovine liver (width: 15 mm, depth: 15 mm, height: 50 mm), MRE using 1-T MR microscope (MR-MICRO, MRTecnology) was used (Table 39.4).

Table 39.2 Multi-frequency MRE imaging parameters by 3-T MRI
Full size table
Table 39.3 Multi-frequency MRE imaging parameters by 0.3-T MRI
Full size table
Table 39.4 MRE imaging parameters by 1-T MR microscope
Full size table

To evaluate the temporal changes, the phantom designed to have a storage modulus of 3 kPa was examined by MRE for the duration of 1 year.

1.2 Ultrasound-Based Shear-Wave Speed Measurement on a Highly Viscous Embedded Phantom

The PAAm gel is composed of a three-dimensional network polymer and a large amount of liquid. The storage modulus (stiffness) of the PAAm gel depends mainly on the quantity of acrylamide. Additionally, the density of the three-dimensional network polymer depends mainly on the quantity of cross-linker. The loss modulus (viscosity) depends mainly on the ratio of water and glycerin. To make compatible phantoms for MRE and USE, the aluminum oxide powder was added to the PAAm gel for the scatterer.

A highly viscoelastic embedded phantom was measured with US-based shear-wave elastography (SWE). The phantom composed square soft part (background part; width: 130 mm, depth: 130 mm, height: 160 mm) and embed two cylindrical hard parts (embedded part; length: 130 mm, diameter: 10 mm and 20 mm). The weight percent of acrylamide in the embedded part is 1.5 times higher than the background part. In addition, we have created a homogeneous phantom that is content the same as the embedded part.

The SWS was measured with virtual touch quantification (VTQ) and VTIQ (Table 39.5). VTQ provides only single-point SWS measurement, and VTIQ provides two-dimensional color-coded SWE (2D SWE), which displays 2D color velocity maps and allows for multiple measurements to be obtained. The stiffness is proportional to squaring of the SWS.

Table 39.5 SWE measurement parameters
Full size table

We measured the embedded parts, and the background parts were 5 mm apart from the outline of the embedded ones. VTIQ measurements were repeated 4 points on the same depth and three times at each part, VTQ measurements were repeated five times at each part, and the mean value and SD of the SWS were calculated. Reference values of the embedded part were measured in a homogenous phantom made with the same material component. The reference value of the background part was measured at a deeper area of the embedded phantom.

2 Results

2.1 Development of a Tissue-Mimicking Viscoelastic Phantom for Quantitative Assessment of MRE

Figure 39.3 shows the stiffness (square root of the sum of squares of the storage and loss modulus) obtained with MRE, the rheometer, and USE. The stiffness of the phantom and bovine liver increased with frequency. Figure 39.4 shows the change in the mechanical properties of the phantom over a year. The change in the storage and loss modulus during the 1-year period was within ±3%.

Fig. 39.3
figure 3

Frequency characteristics of gel phantoms and in-vitro bovine liver

Full size image
Fig. 39.4
figure 4

Changes in the mechanical properties of high-viscosity phantom over time

Full size image

2.2 Ultrasound-Based Shear-Wave Speed Measurement on a Highly Viscous Embedded Phantom

Table 39.6 shows the SWS in the embedded part and background part. In the background part, the SWS was equivalent to the reference value. The SWS of the embedded part with a diameter of 20 mm in a highly viscous phantom was measured accurately with the SWE; however, with a diameter of 10 mm was lower than the reference value. Figure 39.5 shows the B-mode image and the VTIQ image around the embedded part with a diameter of 10 mm. On the VTIQ image, the embedded part was demarcated from the background; however, the border was not sharp. In addition, the embedded part on the VTIQ was visualized to be larger than the part on the B-mode image. This phantom has the potential to be used as a quality control phantom to mimic living tissues.

Table 39.6 Results of SWS measurement
Full size table
Fig. 39.5
figure 5

Image of the phantom containing an embedded part with a diameter of 10 mm. (a) B mode image. (b) VTIQ image

Full size image

3 Conclusion

The purpose of this study was to confirm the frequency characteristics of the developed phantom with commercial USE system and the developed MRE system at multi-scale and multi-frequency. We confirmed that the phantoms are in good agreement with a physical model of the liver, and the developed phantoms are considered effective for the quantitative assessment of the MRE and USE system.

Furthermore, an MRE system using an MR microscope and clinical MRI was developed. We succeeded in performing measurements at multiple frequencies. These developed systems allow quantitative multi-scale and multi-parameter imaging of diseased tissues.