|Ph.D Student||Zurakhov Grigoriy|
|Subject||Improvement of Cardiac Ultrasound Imaging, Utilizing Novel|
|Department||Department of Biomedical Engineering||Supervisors||Professor Emeritus Dan Adam|
|Dr. Zvi Friedman|
|Professor Emeritus Eitan Kimmel|
|Full Thesis text|
Increasing the frame rate of ultrasonic imaging without significantly compromising the quality is one of the most important challenges in cardiac ultrasound. Higher temporal resolution may allow for more accurate visualization of the endocardial-epicardial sequential contraction and more precise speckle tracking echocardiography during stress tests or for patients with a high heart-rate. The frame rate of medical ultrasound imaging systems depends on several factors - penetration depth, the speed of sound through the tissue and the number of transmit (TX) events per frame.
This work is focused on three methods for increasing ultrasound imaging frame rate, Multi-Line Transmission (MLT), Multi-Line Acquisition (MLA) and their combination. In MLT the ultrasound beam is simultaneously focused at a number of directions instead of being focused at a single direction as in the case of a Single Line Transmission (SLT), thus providing a frame-rate improvement by a factor of the number of simultaneously focused beams. This transmission pattern introduces image artifacts known in the literature as cross-talk artifacts. In MLA, a wide transmission beam spans several receive lines, which comes at the cost of a deterioration of the lateral resolution of the ultrasound image. An MLA scan also induces block-like artifacts in the image, which requires proper compensation, for example by a method such as Synthetic Transmit Beamforming (STB). Thus, while using those methods, there is usually a trade-off between the achieved frame-rate improvement and deterioration in the image quality.
In recent years, several adaptive (data dependent) beamforming techniques were applied to medical ultrasound imaging. Those techniques were shown to provide superior image quality to data independent techniques. The main goal of the current study was to further improve the image quality of the cardiac ultrasound images acquired in high frame rate, employing MLT and/or MLA, with modified versions of low complexity adaptive (LCA) beamforming and filtered delay multiply and sum (F-DMAS) beamforming. The current study was performed using a true MLT and/or MLA data, acquired by two ultrasound systems. The results achieved while using an adaptive apodization demonstrate an improvement in image quality, manifested in the reduction of the MLT crosstalk artifacts and improvement of the definition of cardiac structures over the data-independent beamforming with MLT. Moreover, improved results in terms of the lateral resolution were achieved as compared to MLA beamformed with delay and sum (DAS) while the achieved contrast ratio was even better than in case of a single line acquisition (SLT -SLA). Finally, by constructing synthetic transmit beams and performing F-DMAS, we were able to achieve a frame-rate improvement by a factor of 2 over MLT-SLA F-DMAS version, simultaneously improving the contrast of the images as compared to the SLT-SLA beamformed with DAS. Thus, we achieved a 16 fold frame-rate improvement demonstrating a potential to achieve ultra-fast (~391 Hz) high-contrast image.