Pulse Inversion Based Ultrasonic Nonlinear Imaging

• Main Authors: Che-Chou Shen, 沈哲州
• Other Authors: 李百祺
• Format: Others
• Language: en_US
• Published: 2004
• Online Access: http://ndltd.ncl.edu.tw/handle/48305581291641830014

博士 === 國立臺灣大學 === 電機工程學研究所 === 93 === The purpose of this dissertation is to investigate various issues in pulse inversion (PI) based ultrasonic nonlinear imaging. In PI technique, two phase-inverted transmissions are required for each beam. The linear signal is cancelled by summing both echoes together and the nonlinear signal remains in the sum. Image contrast can be significantly improved by the PI technique in ultrasonic nonlinear imaging. The first topic in the dissertation is spectral leakage in nonlinear tissue imaging. Conventional nonlinear imaging suffers from contrast degradation due to spectral leakage when non-negligible harmonic components are present prior to propagation. The leakage signal cannot be filtered out from the received signal due to spectral overlap with the nonlinear signal of interest. Our study is the first complete report in the literature to investigate issues related to spectral leakage in nonlinear tissue imaging. Effects of transmit signal, including envelope and bandwidth, are investigated. Nonlinearity of imaging system and tissue inhomogeneities are also discussed. It is found that the image contrast significantly decreases when spectral leakage is present. The PI technique can restore the image contrast by canceling the leakage signal in the sum, and the trade-off between axial resolution and image contrast is also avoided. Despite of its advantages, the PI technique suffers from frame-rate reduction and potential motion artifacts. Motion artifacts may be critical when there is relative movement between the probe and the imaged tissue during the two firings. In this case, the linear signal increases because of incomplete cancellation in the sum. Effects of motion artifacts on signal intensity and image quality in nonlinear tissue imaging are also studied in the dissertation. It is shown that filtering is still necessary in PI imaging to suppress the uncancelled fundamental signal. Residue spectral leakage also degrades image contrast in the presence of motion artifacts. In the third part of the dissertation, we propose a new PI-based imaging method for contrast nonlinear imaging by using signals in the fundamental band. This method is also known as the PI fundamental imaging. Compared to that obtained using either conventional fundamental imaging or second-harmonic imaging, the contrast-to-tissue ratio (CTR) is significantly improved in PI fundamental imaging. However, motion compensation is essential because the CTR decreases rapidly due to uncancelled tissue fundamental signal when the tissue moves. In addition to its clinical applications, PI fundamental imaging can also be extended to molecular imaging when microbubbles are used as molecular probes. In the appendix, additionally spatial covariance analysis based on the van Cittert-Zernike theorem is also provided for tissue harmonic imaging. It is suggested that the linear signal at lower frequency should be used for effective correction of phase aberration in tissue harmonic imaging.