Ultrasound Multi-angle Plane Waves Ccompounding For Blood Flow Velocimetry Of The Carotid Artery

Cardiovascular and cerebrovascular diseases caused by atherosclerosis show high mortality and disability,which threaten the lives and health of millions of people globally.Hemodynamic parameters based on the blood flow velocity profile(BFVP)are closely related to the progression of atherosclerosis.It is of great significance to accurately measure the BFVP for the prevention,early diagnosis,and risk assessment of cardiovascular and cerebrovascular diseases.Ultrasound diagnostic technology is widely used in clinical detection of blood flow velocity due to its advantages such as non-destructive,non-invasive,and low costs.There are two types of ultrasonic velocimetry,namely,speckle tracking(ST)and color flow imaging(CFI)to estimate the blood flow velocity on the two-dimensional scanning plane.However,ultrasound pulses are transmitted line-by-line in the traditional focused ultrasound.The inter-frame speckle decorrelation caused by the low frame rate(typically one hundred hertz)degrades the performance of the ST method,and the small number of pulses at each spatial position limits the detectable speed range of the Doppler CFI.Through a full aperture,ultrasound plane wave imaging is able to obtain a whole frame of radio frequency(RF)echo signal within a single transmission,thereby increasing the frame rate to more than 20,000 Hz.Moreover,the multi-angle plane wave(MPW)coherent compounding can improve the imaging quality and signal-to-noise ratio while increasing the frame rate.The advantages mentioned above show that the emergence of ultrasonic plane waves provides a new technology for measuring BFVP.However,the current development of ultrasound plane wave imaging still faces many challenges,including:(i)The lack of systematic research on the setting of key parameters such as the number and interval of angles of MPW;(ii)The immature compounding of multi-frame RF echo signal leads to the double loss of frame rate and PRF;(iii)Existing angle sequences are still insufficient and need further improvements;(iv)The motion of blood flow also negatively influences the accurate image compounding and BFVP measurement.To solve the above problems,this thesis has carried out a series of researches on ultrasonic blood flow velocity measurement based on the MPW coherent compounding.The main contents and contributions are as follows:1.For MPW coherent compounding,an increase in the number of angles can generally improve the imaging quality,while reducing the compounded frame rate.In addition,the larger the used interval of inclining angles is,the more the random noise is reduced(i.e.the higher the signal-to-noise ratio is),but at the expense of reduced compounded focal area.In the present study,the performance of MPW coherent compounding for blood flow velocity detection has been assessed based on the Field II platform to explore the best parameter settings for detecting the BFVP.First,two kinds of common carotid artery(CCA)dynamic models(horizontal and inclined models)are built with their known BFVPs being set as the standard parabola.Then,MPWs with the number of angles from 3 to 11 and the interval of angles from 0.5 to 16°are used to scan the horizontal CCA model to generate the sequence of the compounded images.The BFVP is measured by leveraging the ST method,and then compared with the theoretical velocity profile(TVP).The best parameters of MPW are finally determined according to the minimum normalized root mean squared error(NRMSE)between the measured BFVP and TVP.Similarly,the sequence of RF echo signals of the inclined CCA is acquired using MPWs with the angle interval from 0.5°to 20°,and the best parameters of MPW are determined by minimizing the NRMSE between the Doppler CFI-based BFVP and TVP.2.Due to the velocity gradient of BFVP,the blood flow speckle decorrelation rate at each radial position is different.It brings difficulties to the traditional ST method(that uses two adjacent frames of B-mode images)to balance the measurement accuracy of fast and low velocities on the BFVP.In this work,an optimum ST blood flow velocimetry based on the MPW coherent compounding is proposed to improve the measurement accuracy of the BFVP,especially for low velocities near the vessel walls.In the proposed method,the images as far as possible from the reference frame while possessing the blood speckle correlation larger than a given correlated threshold are chosen as the optimum comparison frames from the sequence of the compounded images.The correlated threshold is related to the size of the kernel block being tracked,and is determined in the in-vitro experiment based on a vascular-mimicking phantom.Based on the phantom,the in-vitro experiment using the blood-mimicking fluid confirms that the accuracy and stability of the optimal ST method are higher than those of the traditional ST method.Furthermore,in-vivo experiments using rabbit iliac arteries also verify the clinical feasibility of the proposed method.3.For ultrasonic Doppler velocimetry,the frequency of the Doppler blood flow signal greater than half of the pulse repetition frequency(PRF)can cause velocity aliasing.There is a tradeoff between the increase of the compounded PRF and the improvement of the image quality.Motivated by the above issue,this thesis proposes an alternated polarity plane wave sequence(APPWS)for reusing the coherent adding to break the equivalence between the transmitting and compounded angles during the coherent compounding process.The proposed method can improve the compounded PRF and thus extend the range of the detectable blood flow velocity.Suppose that the inclined angle of the plane wave isβ,the transmitting angle sequence is{0~?-β、~?+、β~?-β、~?},and the compound angle sequence is{0~?、-β~?+、β~?}or{+β~?-β、~?0、~?},in which the last frame of each compounding will be reused for the next compounding.In the context of ensuring that all compounded sequences belong to APPWS,this method can increase the compounded PRF by 50%,thereby increasing the maximum detectable blood flow velocity by 50%.The Field II simulations based on CCA models as well as the in-vitro experiments based on the Shelly vascular-mimicking phantom show that the reused coherent compounding of APPWS realizes the higher accuracy and better stability without the presence of aliasing for measuring fast flow velocities.4.Aiming at addressing the problem in classical plane wave angle sequences,i.e.the virtual focus formed by sidelobe rotation that reduce the lateral resolution and the angle asymmetry,a centrosymmetric polarity plane wave sequence(CPPWS)is proposed,and then combined with the reused coherent compounding to the Doppler CFI estimation.The alternate variations of positive and negative inclined angle in CPPWS can effectively avoid the sidelobe rotation.Moreover,the angle arrangement is symmetrical,so that the proposed sequence cooperating with the reused coherent compounding can improve the compounded PRF.The simulation results of point targets with radial motions show that the reused coherent compounding of the CPPWS improves the imaging performance significantly.Furthermore,the experimental results based on the vascular-mimicking phantom and human carotid arteries along the out-of-plane direction also verify the effectiveness and clinical feasibility of the CPPWS.5.In the two-dimensional scanning plane,the blood flow velocity is different and time-varying at each spatial position,which results in motion artifacts in the compounded RF echo signals of MPWs and affects the accuracy of velocity measurement.In this thesis,the spatial-temporal motion compensation(Mo Co)method for MPW coherent compounding is proposed to eliminate motion artifacts from the compounded RF signals.The Mo Co method takes into account the spatial difference of blood flow velocities and the time continuity of the compounded sequence simultaneously.Before applying the Mo Co,the two-dimensional scanning plane is divided into several local areas,and the phase shifts of the MPW echo signals are calculated based on the matching cost function in each local area.With the calculated phase shifts,the Mo Co is performed to align the MPW echo signals to the middle time of the MPW scanning duration.The performance of the proposed method based on two matching cost functions,namely,the maximum normalized cross correlation(NCC)and the minimum sum square difference(SSD),is evaluated.In simulations,the experiments based on static point target show that the spatial-temporal Mo Co method can eliminate the adverse effect of the inclining angles on the coherence of the echo signals.The experiments based on the moving point target show that the method can eliminate the motion artifacts in the compounded image.6.For moving target with low velocities,the phase shift between any adjacent scans is too small to be estimated.This hampers the direct use of the Mo Co method.In order to reduce the quantization errors of the phase shift estimation,an improved spatial-temporal Mo Co method based on sub-sample estimation(Mo Co-Sub)is proposed.During the calculation of the phase shift of two signals,sub-sample displacement estimation refers to interpolating the results of the matching cost function,and then finding the optimal matching value among the interpolation results.In doing so,the lowest unit of phase shift estimation is lower than the physical distance between two sampling points,so that the quantization error is reduced.The proposed method is verified on the simulation of moving point targets as well as the experiments on blood-mimicking fluid and human carotid arteries.It is demonstrated that the Mo Co-Sub method possesses higher BFVP measurement accuracy than the original Mo Co method.With the goal of improving the accuracy of BFVP measurement,this work has studied and developed a number of ultrasonic blood flow velocimetry techniques based on the MPW coherent compounding.They have provide a new foundation for the analysis of hemodynamic conditions,by which more accurate diagnostic information can be obtained to promote the prevention,early diagnosis,disease course monitoring and risk assessment of atherosclerosis.