Date of Award
Mechanical, Automotive, and Materials Engineering
Adaptive focusing, Image reconstruction, Transcranial, Ultrasound
CC BY-NC-ND 4.0
Transcranial ultrasound phased array imaging through the thick human skull bone and outside of the temporal acoustic windows is investigated through the current study. The significant sound speed discrepancy between the skull bone and the brain tissue introduces phase aberration and refraction to the propagating wave fronts and consequently causes beam defocusing and image degradation. A non-invasive adaptive focusing technique is presented to compensate for the aberration effect of the skull bone in the transmission mode. For this purpose, it is necessary to extract the profiles of the skull bone through a preliminary step. The feasibility of 3-D skull profile extraction using an ultrasound matrix array is investigated for the first time. Two methods, multi-lag phase delay (MLPD) estimation and modified space alternating generalized expectation maximization (SAGE) are proposed to extract the map of the skull boundaries. Taking advantage of numerical modeling the method exploits multiple virtual acoustic sources embedded in the intended focal points behind the skull phantom and numerically tracks the propagating wave fronts through the heterogeneous medium in a finite difference framework. Using the computed arrival times at the coordinates of each element a new time delay set is generated and introduced to the transducer elements for focusing and steering through the skull bone. Numerical and experimental results show that the quality of focus is significantly improved through the presented procedure. To characterize the distorting effects of the skull barrier on transcranial images the point spread function (PSF) of the imaging system is numerically modeled with and without the presence of skull. A method for refraction correction of transcranial passed array images is proposed. Exploiting the gradient information of the computed time-of-flights over the interrogated area, the algorithm could successfully track the ray propagation paths from each focal point to the active aperture center. Dynamic focusing is then achieved along the estimated paths. The reconstructed images exhibit a 40% enhancement in contrast and a 38% increase in detection rate. Finally, a multiscale-based method is presented to suppress the speckle noise and detect the discontinuities and objects boundaries in ultrasound images, simultaneously.
Hajian, Mehdi, "Reconstruction and Analysis of Ultrasound Images for Transcranial Ultrasound Applications" (2016). Electronic Theses and Dissertations. 5825.