Ultrasound Measurement Of Blood Flow Velocity In The Coding Techniques

Ultrasound Doppler blood flow measurement system (UDBFMS) plays a significant role in medical ultrasound diagnostic devices. The precision of flow measurement, signal-to-noise ratio (SNR), sensitive of system detection are always the hot topics in the research of medical ultrasound. Recently, coding technique has been widely used in medical ultrasound images for its advantages in improving the image quality and increasing the penetration depth and sensitivity. But the coding technique in UDBFMS still lacks progress.In the dissertation, several key issues of coding technique in UDBFMS are addressed. Especially, coded modulation and pulse compression are studied in details. A new coding and decoding method based on improved m code and a new pulse compression method based on weighted spiking filter are proposed for coding technique. Theoretical analysis and simulations are carried out to validate the performance improvements in UDBFMS. It is shown that these two new methods are efficient ones to improve the precision and sensitivity of Doppler flow measurement.In the pseudo-random code-based UDBFMS, a novel coding and decoding method is proposed to achieve better correlation properties and remove all clutters at distant positions. This method is better than existing ones in removing clutters in nearby non-desired region. The results of the simulation studies, in which Faran's spherical elastic scatter model is used to enhance the simulation speed, show that the flow velocity in the desired region can be measured with good precision when using this novel coding and decoding method. In a cardiac cycle simulation of pulsatile flows, mean absolute error (MAE) and root-mean-square (RMS) error of this new method are more than two times less than those of Shiozaki method and are approximately one-third of those of standard m-code method.In the coded excitation pulse-based UDBFMS, a weighted spiking filter method is proposed to give different weights to the filter outputs with the objective of minimizing the weighted RMS error of filter output. This method can render more precise filter output at the peak value and better velocity profile measurement for the system. The simulations on steady streams and pulsatile flows are carried out to compare our weighted spiking filter with traditional inverse filter and spiking filter. The results show that weighted spiking filter has the best velocity measurement values with lower RMS error and stronger noise robustness than inverse filter and spiking filter, while the performance of pulse compression is independent on flow velocity. The impact of filter length on velocity measurement is also studied by using simulations. It is shown that the optimum filter length in our simulation environment is 20 to 45.