Today ultrasound (US) echography is a technique widely used for diagnosis of
many diseases. This success stems from the fact that it is non-invasive, simple
to implement and relatively low cost compared to other imaging modalities. In
recent years, the apparition of ultrasound contrast agent has developed a new
imaging mode: the tissue harmonic imaging (THI). Indeed, the nonlinear
propagation of the pressure wave into the medium produces harmonics that can be
imaged. THI has also been used without contrast agent due to its better
resolution compare to classical ultrasound imaging. However, in the US
community, no tools have been developed to simulate realistic images of such
situation. Such a tool opens new perspectives in terms of simulation and
ultrasound method design and improvements.
In the proposed seminar, I will present a new tool to simulate realistic
nonlinear ultrasound images. This tool is composed of two separate bricks.
First, a nonlinear propagation simulator, valid with media with inhomogeneity
in
the nonlinearity, is proposed. An angular spectrum method has been used to
reduce the derivative order of the propagation equation and proposed a simpler
mathematical resolution. Then, the method has been implemented on a CPU and a
GPU. Then, an image reconstruction algorithm is provided to simulate the final
nonlinear image and future applications are discussed.

At the end of the presentation, I will briefly present the current work I am
doing at CMP on the segmentation of mouse neuromuscular axonal projection
fibers.