TY - JOUR
T1 - Multi-plane ultrafast compound 3D strain imaging
T2 - Experimental validation in a carotid bifurcation phantom
AU - Fekkes, Stein
AU - Saris, Anne E.C.M.
AU - Menssen, Jan
AU - Nillesen, Maartje
AU - Hansen, Hendrik H.G.
AU - de Korte, Chris L.
PY - 2018/4
Y1 - 2018/4
N2 - Strain imaging of the carotid artery (CA) has demonstrated to be a technique capable of identifying plaque composition. This study assesses the performance of volumetric strain imaging derived from multi-plane acquisitions with a single transducer, with and without displacement compounding. These methods were compared to a reference method using two orthogonally placed transducers. A polyvinyl alcohol phantom was created resembling a stenotic CA bifurcation. A realistic pulsatile flow was imposed on the phantom, resulting in fluid pressures inducing 10% strains. Two orthogonally aligned linear array transducers were connected to two Verasonics systems and fixed in a translation stage. For 120 equally spaced elevational positions, ultrasound series were acquired for a complete cardiac cycle and synchronized using a trigger. Each series consisted of ultrafast plane-wave acquisitions at 3 alternating angles. Inter-frame displacements were estimated using a 3D cross-correlation-based tracking algorithm. Horizontal displacements were acquired using the single probe lateral displacement estimate, the single probe compounded by axial displacement estimates obtained at angles of 19.47 and -19.47 degrees, and the dual probe registered axial displacement estimate. After 3D tracking, least squares strain estimations were performed to compare compressive and tensile principal strains in 3D for all methods. The compounding technique clearly outperformed the zero-degree method for the complete cardiac cycle and resulted in more accurate 3D strain estimates.
AB - Strain imaging of the carotid artery (CA) has demonstrated to be a technique capable of identifying plaque composition. This study assesses the performance of volumetric strain imaging derived from multi-plane acquisitions with a single transducer, with and without displacement compounding. These methods were compared to a reference method using two orthogonally placed transducers. A polyvinyl alcohol phantom was created resembling a stenotic CA bifurcation. A realistic pulsatile flow was imposed on the phantom, resulting in fluid pressures inducing 10% strains. Two orthogonally aligned linear array transducers were connected to two Verasonics systems and fixed in a translation stage. For 120 equally spaced elevational positions, ultrasound series were acquired for a complete cardiac cycle and synchronized using a trigger. Each series consisted of ultrafast plane-wave acquisitions at 3 alternating angles. Inter-frame displacements were estimated using a 3D cross-correlation-based tracking algorithm. Horizontal displacements were acquired using the single probe lateral displacement estimate, the single probe compounded by axial displacement estimates obtained at angles of 19.47 and -19.47 degrees, and the dual probe registered axial displacement estimate. After 3D tracking, least squares strain estimations were performed to compare compressive and tensile principal strains in 3D for all methods. The compounding technique clearly outperformed the zero-degree method for the complete cardiac cycle and resulted in more accurate 3D strain estimates.
KW - 3D strain estimation
KW - Bifurcation
KW - Dual probe acquisitions
KW - Plane wave ultrasound
KW - Polyvinyl alcohol phantom
KW - plane wave ultrasound
KW - bifurcation
KW - dual probe acquisitions
KW - polyvinyl alcohol phantom
UR - http://www.scopus.com/inward/record.url?scp=85045770314&partnerID=8YFLogxK
U2 - 10.3390/app8040637
DO - 10.3390/app8040637
M3 - Article
AN - SCOPUS:85045770314
SN - 2076-3417
VL - 8
JO - Applied Sciences
JF - Applied Sciences
IS - 4
M1 - 637
ER -