TY - GEN
T1 - CNN and back-projection
T2 - Medical Imaging 2019: Ultrasonic Imaging and Tomography
AU - Anas, Emran Mohammad Abu
AU - Cheng, Alexis
AU - Seifabadi, Reza
AU - Wu, Yixuan
AU - Aalamifar, Fereshteh
AU - Wood, Bradford
AU - Rahmim, Arman
AU - Boctor, Emad M.
N1 - Publisher Copyright:
© 2019 SPIE.
PY - 2019
Y1 - 2019
N2 - The potential of ultrasound tomography has been noticed to quantify the tissue acoustic properties for advanced clinical diagnosis. However, the location of most of the human anatomies limits the tomography for a few angles that leads the reconstruction as a more challenging problem. In this work, a deep convolutional neural networks- based technique is presented to estimate the speed of sound of tissue from a limited angle projection data. The underlying concept is based on filtered back projection technique, where the convolutional neural network is used to model the high-pass filter before the back projection. Moreover, we use a post convolutional neural network to suppress the artifacts generated due to the limited angle tomography. We train the network from a set of simulation experiments; on the test set consisting of 1,750 simulation experiments, we achieve an average mean absolute error of 2.1% in predicting the speed of sound map.
AB - The potential of ultrasound tomography has been noticed to quantify the tissue acoustic properties for advanced clinical diagnosis. However, the location of most of the human anatomies limits the tomography for a few angles that leads the reconstruction as a more challenging problem. In this work, a deep convolutional neural networks- based technique is presented to estimate the speed of sound of tissue from a limited angle projection data. The underlying concept is based on filtered back projection technique, where the convolutional neural network is used to model the high-pass filter before the back projection. Moreover, we use a post convolutional neural network to suppress the artifacts generated due to the limited angle tomography. We train the network from a set of simulation experiments; on the test set consisting of 1,750 simulation experiments, we achieve an average mean absolute error of 2.1% in predicting the speed of sound map.
KW - Altered back-projection
KW - Convolutional neural networks
KW - Speed of sound
KW - Ultrasound tomography
UR - http://www.scopus.com/inward/record.url?scp=85066628161&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85066628161&partnerID=8YFLogxK
U2 - 10.1117/12.2513043
DO - 10.1117/12.2513043
M3 - Conference contribution
AN - SCOPUS:85066628161
T3 - Progress in Biomedical Optics and Imaging - Proceedings of SPIE
BT - Medical Imaging 2019
A2 - Byram, Brett C.
A2 - Ruiter, Nicole V.
PB - SPIE
Y2 - 17 February 2019 through 18 February 2019
ER -