deletions | additions       diff --git a/Introduction.tex b/Introduction.tex  index 2155345..705cb9b 100644  --- a/Introduction.tex  +++ b/Introduction.tex 
...
When a laser beam of sufficient energy is incident on a medium, the absorption of the electromagnetic radiation leads to an increase of the local temperature. Due to thermal effects, displacements occur in the medium. These displacements can then propagate as elastic waves. Elastic waves are separated in two components in a bulk: compression waves, corresponding to a curl-free propagation; and shear waves, corresponding to a divergence-free propagation \cite{aki2002quantitative}. Measures of the compression and shear waves is notably used as a method of inspection to reveal potential cracks in a solid such as a metal. In a medical context, induction of compression waves by laser has been studied with the development of photoacoustic imaging \cite{Xu_2006}, \cite{22442475}. Elastic waves used in photoacoustic imaging are typically of a few megahertz: at this frequency, in a soft tissue, shear waves are quickly attenuated, typically over a few microns in soft tissues, microns,  so only compression waves can propagate over a few centimeters.   Shear waves have however nevertheless  drawn an increasing interest in medical imaging, with the development of shear wave elastography techniques for the last two decades \cite{muthupillai1995magnetic}, \cite{sandrin2002shear}. As its names indicates, this term covers the techniques used to measure or map the elastic properties of biological tissues using shear wave propagation. The shear modulus, directly proportional to Young's modulus in soft tissues, varies of several orders of magnitude between different tissues in human body and potentially offers a good contrast. As a shear wave propagates in an organ at a speed proportional to the square root of the shear modulus, measuring its speed throughout the organ allows to compute the shear modulus of the tissue. Shear waves can be induced by an external vibrator \cite{muthupillai1995magnetic}, a focused acoustic beam \cite{sarvazyan1998shear}, \cite{11937286}, the Lorentz force \cite{16051039}, \cite{grasland2014elastoEMarticle}, or natural body displacements \cite{gallot2011passive}. \cite{gallot2011passive}, \cite{Zorgani_2015}, \cite{Weaver_2012}.  Shear wave elastography techniques have been successfully applied for the detection of various pathologies in organs such as the liver \cite{sandrin2003transient}, the breast \cite{goddi2012breast}, \cite{sinkus2005viscoelastic}, the prostate \cite{cochlin2002elastography}, \cite{12878247}, and the eye cornea \cite{tanter2009high}. Very recently, Li et al. have used induce surface acoustic wave induced with a laser to assess bladder wall elasticity \cite{25574440} and cornea elasticity \cite{22627517}. In this study, we show that shear waves can be induced in soft tissues by a laser beam. We also propose a model for the underlying physics. We finally applied the technique in a biological tissue to evaluate its potential application in shear wave elastography.  In the first experiment, illustrated by Figure \ref{Figure1}, we used a Q-switch Nd:YAG laser (EverGreen 200, Quantel, Les Ulis, France), which produced a pulse of energy $E$ = 200 mJ at a central wavelength of 532 nm during 10 ns in a 5 mm diameter circular beam of 5 mm diameter (so that section $S$=20 mm$^2$). beam.  We defined Z as the laser beam axis and the laser impact location on the medium as the origin of coordinates (0,0,0). The laser beam was absorbed in a 4x8x8 cm$^3$ tissue-mimicking black mat phantom made of water and of 5\% polyvinyl alcohol, 1 \% black graphite powder and 1\% salt. A freezing/thawing cycle was applied to stiffen the material to a value of 15$\pm$5 kPa \cite{17375819}.The laser is absorbed in the medium with an exponential decay of the optical intensity $I(z)=I_0 \exp(- \gamma z)$ along medium depth $z$, where $I_0=\frac{1}{S}\frac{d E}{dt}$ is the incident intensity distribution at the surface (the reflection on the black mat material being neglected) and $\gamma$ the absorption coefficient of the medium.  The absorption coefficient $\gamma$ can be estimated by calculating the skin depth $\delta = (\pi \sigma \mu_r \mu_0 \nu)^{\frac{1}{2}}$, where $\sigma$ is the electrical conductivity of the medium, $\mu_r \mu_0$ its permeability and $\nu$ the frequency of the radiation. Substituting $\sigma$ = 1 S.m$^{-1}$, $\mu_r \mu_0 \approx 4 \pi \times 10^{-7} H.m^{-1}$ and $\nu$ = 532 nm, the skin depth for our medium is about 47 $\mu$m. We have validated experimentally this value by measuring the fraction of light which go through different thicknesses of the medium (respectively 0, 30, 50 and 100 $\mu$m) with a laser beam power measurement device (QE50LP-S-MB-D0 energy detector, Gentec, Qu\'e bec, QC, Canada). We found respective power of 100\%, 88\%, 71\% and 57\%. An exponential fit indicated that $\gamma^{-1} \approx$ 50 $\mu m$ in our sample, meaning that 63\% of the radiation is absorbed in the first 50 micrometers of the sample.  The absorption of the laser beam by the medium gives then rise to an absorbed optical energy $q = \gamma I$. Assuming that all the optical energy is converted to heat, a local increase of temperature appears. Temperature distribution $T$ in absence of convection and of phase transition can be computed using the heat equation:  \begin{equation}  \frac{\partial T}{\partial t} = \kappa \nabla ^2 T + \frac{q}{\rho C}  \label{eq:eqChaleur}  \end{equation}  where $\rho$ is the density, $C$ the heat capacity and $\kappa$ the thermal diffusivity. The thermal diffusion path, equal to $\sqrt{4\kappa \tau}$, with $\tau$ = 10 ns the laser emission duration and $\kappa$ = 1.43 10$^{-7}$ m$^2$.s$^{-1}$ for water \cite{Blumm_2003}, is approximately equal here to 80 nm. As $\gamma^{-1} \gg \sqrt{4\kappa t}$, propagation of heat is negligible during laser emission, so that equation \ref{eq:eqChaleur} can be simplified as $\frac{\partial T}{\partial t} = \frac{q}{\rho C}$. Substituting the experimental parameters lead to a maximum increase of temperature of 12 K for maximum laser energy (200 mJ).  This local increase of temperature can lead a local dilatation of the medium occurs. We suppose that the medium is homogeneous and isotropic, and as the depth of absorption is small compared to the beam diameter, we adopt a 1D model. The stress $\sigma_{zz}$ is the sum between the axial strain component and the thermal expansion component \cite{scruby1990laser}:  \begin{equation}  \sigma_{zz} = (\lambda + 2 \mu) \frac{\partial u_z}{\partial z} - 3(\lambda + \frac{2}{3}\mu) \frac{\alpha E}{\rho C S \zeta}  \label{eq:stressThermo}  \end{equation}  where $\lambda + 2 \mu$ is the P-wave modulus and $\lambda + \frac{2}{3}\mu$ the bulk modulus with $\lambda$ and $\mu$ respectively the first and second Lamé's coefficient, $\alpha$ is the thermal dilatation coefficient and $\zeta$ the average depth of laser beam absorption. In the absence of external constraints normal to the surface, the stress across the surface must be zero, i.e. $\sigma_{zz} (z=0) = 0$, so that equation \ref{eq:stressThermo} can be integrated, giving a displacement $u_z$ from the surface:  \begin{equation}  u_z = \frac{(3\lambda + 2\mu)}{(\lambda + 2\mu)} \frac{\alpha E \zeta}{\rho C S \zeta}  \label{eq:deplThermo}  \end{equation}  As in a biological soft tissues, $\mu \ll \lambda$, the displacement $u_z$ can be approximated as $\frac{3 \alpha E}{\rho C S}$. Taking as an order of magnitude $\alpha$ = 70.10$^{-6}$ K$^{-1}$ (water linear thermal dilatation coefficient), $E$ = 200 mJ, $\rho$ = 1000 kg.m$^{-3}$ (water density), $C$ = 4180 kg.m$^{-3}$ (water calorific capacity) and $S$ = 20 mm$^2$, we obtain a displacement $u_z$= 0.5 $\mu$m. This value is slightly smaller than the typical ultrasound elastography resolution (5--10 $\mu$m \cite{22545033}). In a 3D model, displacements along X and Y axis would also occurs, as the local expansion acts as dipolar forces parallel to the surface.  If the laser beam is focused, the local increase of temperature can vaporize a part of the surface of the medium.  In both cases, absorption of the laser by the phantom leads to a local displacement which can propagate as elastic waves in the medium. In both cases, shear wave can occur because of the limited size of the source.  To observe the resulting  shear waves, the medium was scanned with a 5 MHz ultrasonic probe made of 128 elements connected to a Verasonics scanner (Verasonics V-1, Redmond, WA, USA). The probe was used in ultrafast mode \cite{bercoff2004supersonic}, acquiring 1500 ultrasound images per second. Due to the presence of graphite particles, the medium presented a speckle pattern on the ultrasound image. Tracking the speckle spots with an optical flow technique (Lucas-Kanade method) allowed to compute one component of the displacement in the medium (Z-displacement or Y-displacement, depending on the position of the probe on the medium). The laser beam was triggered 10 ms after the first ultrasound acquisition, $t$ = 0 ms being defined as the laser emission.