Research On Rapid Ultrasound Transcranial Blood Flow Imaging Method Based On Divergent Waves

Cerebrovascular disease is one of the most common clinical diseases.The incidence rate is increasing year by year.Studies have shown that cerebrovascular disease is the second leading cause of death in the world,but many cerebrovascular diseases do not show any symptoms during development.Ultrasound imaging can detect the blood flow noninvasively and in real time,and can locate the ischemic area in the large intracranial artery,which enables it to diagnose early cerebrovascular disease,monitor the patient's condition,provide appropriate treatment reference,and track the recovery of patients after prognosis.At present,the methods of transcranial blood flow imaging are all based on the focus line scanning mode,but this emission mode is limited by the frame rate,so it is impossible to accurately estimate the high-speed blood flow.In order to achieve a certain frame rate,it is necessary to reduce the number of scan lines for imaging,but it will narrow the imaging field of vision.Because of the mismatch between the sound speed of skull and soft tissue,the phase distortion of ultrasound will occur during transcranial,resulting in the position of blood flow image is shifted.Therefore,the goal of this study is to obtain transcranial blood flow images with high frame rate,large field of view and accurate location of blood vessels.Aim at the above problems,we use the diverging wave to realize the blood flow imaging with higher frame rate and larger field of view,and use the Singular Value Decomposition(SVD)wall filter to improve the ability of filtering the clutter signal in the limit state,then we introduce two kinds of finite element difference methods(Multi-Template Fast Matching-MFMM and Wavefront extrapolation-EW),they are used to correct the position deviation of blood flow image caused by phase distortion of skull.The main research contents and results of this paper are as follows:(1)the application of two emission modes of Coherent plane-wave compounding and Coherent diverging-wave compounding in blood flow imaging is realized,and the two emission modes are compared by using Field II simulation platform.The horizontal resolution of plane wave mode is better than that of diverging wave mode,and at the maximum deflection angle used in this study,after compounding 5 and 3 angles respectively,the horizontal resolution will not increase withthe increase of the number of angles.The difference between them in contrast is very small,and when the number of array elements is small,the diverging wave mode is better than that of plane wave mode,and the contrast will increase with the increase of the number of angles.The accuracy of blood flow velocity estimation of them is close.The accuracy of blood flow velocity estimation increases with the increase of the number of angles.The diverging wave mode has the advantage of larger imaging field than the plane wave mode.(2)The performance of Infinite Impulse Response-IIR(include three initialization methods),polynomial regression filter and SVD as wall filter are compared by the phantom experiment.The results show that Polynomial Regression has the highest accuracy of blood flow velocity estimation,followed by SVD,they were all achieved over 90%,and SVD shows the ability of filtering the clutter signal in the limit state.(3)MFMM and EW phase correction algorithms are used in divergent wave blood flow imaging.k-Wave simulation experiment and skull experiment are done to verify the correction ability of the two algorithms for image position.The results show that the two algorithms can correct the position of blood flow image in axial and transverse direction.The corrected performance of MFMM is closer to that of skull free imaging,while EW has the advantage in computing time.