A speed of sound map of tissue using ultrasound can be used to diagnose various issues with the tissue while not being as costly as other forms of imaging. An emerging method of generating the speed of sound map from ultrasound signals is through deep learning. As with any deep learning method there are often hyper parameters that need to be tuned for better results. To accomplish this, I used a variation of grid search to find the optimal parameters for the network. The results were compared by calculated the root mean squared error (RMSE) of test data generated either via full wave simulation or straight wave simulation. Upon tuning the optimal learning rates, batch size, and loss function weighting decrease in the RMSE can be observed.
Use this login method if you
don't
have an
@illinois.edu
email address.
(Oops, I do have one)
IDEALS migrated to a new platform on June 23, 2022. If you created
your account prior to this date, you will have to reset your password
using the forgot-password link below.
