Statistical Analysis of Doppler Parameters Using Microbubbles

• Main Authors: Men-Tzung Lo, 羅孟宗
• Other Authors: Jenho Tsao
• Format: Others
• Language: en_US
• Published: 2004
• Online Access: http://ndltd.ncl.edu.tw/handle/48962027881095908098

博士 === 國立臺灣大學 === 電信工程學研究所 === 92 === The ultrasound system for drug delivery should provide the function to recognize the coming medicine carrier and destroy it to release the interior drug. Note that the contrast agent (i.e. microbubble) is adopted as the medicine carrier. Therefore, the goal of this study is to formulate two theorems about bubble detection and cavitation detection for drug delivery. The Doppler domain analysis for echoes backscattered from microbubbles is necessary, since in Doppler domain we only need to separate the microbubbles from the red cells rather than other tissues. In addition, the first step in the procedure for analyzing the Doppler data is to build the statistical model (such as autocorrelation function) of echoes, since the backscattering signals form the distributed microbubbles and red cells are random process. Under band limited transmission, it can be observed that beyond a critical frequency, the theoretical volume backscattering cross-section (power gain) derived from the ensemble average power spectrum of microbubbles decreases with frequency. On the contrary, the volume backscattering cross-section of red cells increases with frequency. The decreasing gain with frequency for distributed microbubbles is proved theoretically to result in the downward shift in estimated Doppler parameters after microbubble injection. If we increase the transmitted bandwidth, the variation in estimated Doppler parameters after microbubble injection will become more apparent, this benefits for bubble reorganization. The variations in Doppler parameters estimated with experimental data are presented to verify the theoretical deviations. Under the narrow band insonification for bubble cavitation, the appearance of the apparent subharmonics is proposed to the indication of cavitation in this study, since the onset threshold of the subharmonics is very high. If the onset threshold of the subharmonics is so high that it is close to the cavitation threshold, the observation time for the subharmonics will be very short; this is proved theoretically to result in the weak correlation of consecutive Doppler signals in this study. Therefore, the estimated Doppler signals for the subharmonics can be used to specify whether the appearance of subharmonics can be used to indicate the cavitation. Our numerical results demonstrated that the onset thresholds are indeed very close to the cavitation thresholds for most bubbles. Our experimental results carried out using the suspension of levoist® and free gas show that the correlation of two consecutive Doppler signals obtained form the apparent subharmonics is very weak. The experimental results also present that the weak correlation of Doppler signals should be accused of the excessively reduced observation time rather than the flow velocity spread.