## Introduction

## Shear Wave Dispersion Imaging

### Theory

*c*at SW frequency

_{s}*ω*is calculated by the Voigt model as shown below [11,12]:

*ρ , μ*, and

*η*are the density, shear elasticity, and shear viscosity of the medium, respectively.

*η*in Eq. (1)-that is, a perfectly elastic tissue is assumed and viscosity is ignored-the SW speed

*c*can be transformed into a simple relationship with shear elasticity:

_{s}### Shear Wave Dispersion Maps

*i*-series, Canon Medical Systems). This technique can be used to estimate the dispersion slope of SW speed versus frequency in order to evaluate changes in tissue viscosity.

*c*(

_{s}*ω*) is as shown below:

*Δø*(

*ω*) is the phase change over the distance traveled

*ΔL*between two measurement locations in the direction of SW propagation. Fourth, the gradient of SW speed is calculated based on the distribution of SW speed versus frequency. The calculated gradient values are then superimposed on the measurement locations to create a dispersion map.

### Shear Wave Dispersion Imaging

*i*-series are as follows. SWD can be activated automatically in SWE mode. The dispersion map shows dispersion slope, which is a parameter directly related to viscosity. The calculated dispersion slope value ([m/sec]/kHz) and its standard deviation are displayed. In SWE quad-view mode, SW speed or SW elasticity (speed map, elasticity map), SW arrival time contours (propagation map), grayscale, and dispersion slope (dispersion map) can be viewed simultaneously (Fig. 3).

## Preclinical Study

_{4}] injected twice a week for 1 week), G2 (CCl

_{4}injected 4 times a week for 1 week), G3 (CCl

_{4}injected twice a week for 6 weeks), and G4 (CCl

_{4}injected twice a week for 10 weeks). G1 and G2 comprised a necroinflammation rat model, and G3 and G4 comprised a fibrosis rat model.

## Clinical Studies

*i*800, Canon Medical Systems) in patients with biopsy-proven nonalcoholic fatty liver disease (NAFLD; n=24). In a multivariable analysis with histologic features as independent variables, fibrosis stage was found to be significantly related to SW speed (P=0.037) and lobular inflammation grade was significantly related to dispersion slope (P=0.022). SW speed was found to be more useful than dispersion slope for predicting the degree of fibrosis, and dispersion slope was found to be more useful than SW speed for predicting the degree of necroinflammation. These findings suggest that performing dispersion slope measurements, as well as SW speed measurements, may lead to more accurate diagnosis, staging, and treatment monitoring of nonalcoholic steatohepatitis (NASH) than is possible using existing imaging methods (Fig. 5).