|dc.description.abstract||In the field of ultrasound imaging, it has been theorized that imaging noise, known as
speckle, is the product of microscopic scatterers and abnormalities within the imaged tissue.
This would result in certain speckling patterns revealing themselves over large datasets,
which could be utilized to identify minuscule lesions within tissues, potentially creating a
method to predict the early formation of tumors. Such a dataset would be difficult to analyze
by hand, but machine learning algorithms could be used to recognize patterns in a effective
manner. As of now, few attempts have been made to utilize machine learning in order to
predict scatterer placement from ultrasound scans.
In order to initiate machine learning, first a computational simulation must be constructed
to consistently and accurately reproduce experimental data. Using Field II, a
MATLAB-based program for ultrasound modelling, simulations were created to replicate
data produced from experimental phantoms made from glass beads and agarose gel. These
simulations were designed to account for bead placement and size, as well as experimental
conditions. Comparisons between simulations and experimental data using statistical analysis
show that ultrasound images can accurately be predicted using computational methods.
With these software programs, it becomes possible to train a machine learning algorithm
to recognize speckling pattern, which may allow for the resolution of previously unresovable