TY - JOUR
T1 - Comparison of statistical models for the detection of uniform reverberant shear wave fields
AU - Miranda, Edmundo A.
AU - Castaneda, Benjamin
AU - Romero, Stefano E.
N1 - Publisher Copyright:
© 2021 IEEE.
PY - 2021
Y1 - 2021
N2 - Reverberant shear wave field elastography (R-SWE) is an image modality that evaluates tissue stiffness by creating a reverberant shear wave field in all directions of the medium. This implementation facilitates the estimation of the shear wave speed (SWS), ergo the viscoelasticity of the tissue. Its feasibility has been validated in breast, liver, corneas and plantar tissue. Nevertheless, the calculus of the SWS is verified as long as a uniform field is generated. Current method uses the coefficient of determination (R2) derived of a curve fitting as a quality parameter of reverberation. In this work, the uniformity phenomenon in reveberant fields is studied through the extraction and analysis of statistical estimators used in electromagnetic fields with their equivalency in acoustic waves to implement machine learning classifiers of uniform reverberant fields such as Logistic Regression, LDA, Lineal SVM and Gaussian SVM. The models detect uniform regions with accuracies of 0.749, 0.746, 0.715 and 0.811, respectively.
AB - Reverberant shear wave field elastography (R-SWE) is an image modality that evaluates tissue stiffness by creating a reverberant shear wave field in all directions of the medium. This implementation facilitates the estimation of the shear wave speed (SWS), ergo the viscoelasticity of the tissue. Its feasibility has been validated in breast, liver, corneas and plantar tissue. Nevertheless, the calculus of the SWS is verified as long as a uniform field is generated. Current method uses the coefficient of determination (R2) derived of a curve fitting as a quality parameter of reverberation. In this work, the uniformity phenomenon in reveberant fields is studied through the extraction and analysis of statistical estimators used in electromagnetic fields with their equivalency in acoustic waves to implement machine learning classifiers of uniform reverberant fields such as Logistic Regression, LDA, Lineal SVM and Gaussian SVM. The models detect uniform regions with accuracies of 0.749, 0.746, 0.715 and 0.811, respectively.
KW - R-SWE
KW - SWS
KW - reverberant field
KW - uniformity
UR - http://www.scopus.com/inward/record.url?scp=85122882304&partnerID=8YFLogxK
U2 - 10.1109/IUS52206.2021.9593570
DO - 10.1109/IUS52206.2021.9593570
M3 - Conference article
AN - SCOPUS:85122882304
SN - 1948-5719
JO - IEEE International Ultrasonics Symposium, IUS
JF - IEEE International Ultrasonics Symposium, IUS
T2 - 2021 IEEE International Ultrasonics Symposium, IUS 2021
Y2 - 11 September 2011 through 16 September 2011
ER -