Adaptive Harmonic Separation Of Ultrasonic Echo Signals And Its Application In Imaging Monitoring Of Tissue Ablation

Harmonic imaging is an imaging technique that receives and utilizes the harmonic information generated by the nonlinear propagation of ultrasonic waves in biological tissues and the nonlinear vibration of microbubbles in contrast media.It is a research hotspot in the field of ultrasonic nonlinear medical diagnosis in recent years.On the one hand,the non-linear propagation of ultrasonic in the tissue can result in the echo of second and higher harmonics,which can improve the contrast and axial resolution of ultrasonic images.On the other hand,the ultrasound contrast agent,as a nonlinear scatterer,can generate second,higher,sub-and ultra-harmonic components in the ultrasonic radio frequency(RF)echo signals.Sub-harmonic and ultra-harmonic imaging can effectively improve the contrast-tissue-ratio(CTR)of ultrasound images.Each order of harmonic imaging possesses its unique characteristics and applicability.It can describe nonlinear changes caused by changes in tissue structure,effectively monitor thermal damage caused by microwave ablation,and expand the scope of clinical diagnosis and improve the diagnostic level.The complete and accurate separation of the harmonic components from the ultrasonic echo RF signal has become the focus of the research on improving the imaging quality.Band-pass filtering and pulse inversion are two commonly used methods for separating the multi-order harmonics from RF echo signals.However,the cutoff frequency,order,and algorithm to realize a band-pass filter have a significant influence on the separation performance.The movement of the transducer and tissue can produce large separation errors in harmonic imaging based on pulse inversion method.In response to the above problems,harmonic separation methods based on adaptive nonlinear signal analysis are proposed in this study to overcome the defects of existing methods and improve the quality of harmonic imaging.In order to improve the monitoring performance of microwave ablation,a harmonic Nakagami parameter imaging method based on adaptive nonlinear harmonic separation is proposed according to the nonlinear changes produced during the heating of microwave ablation tissue.The main contents and innovative achievements of the research are as follows:1.According to the characteristics of spectrum of harmonic components in the ultrasonic echo signal,a novel separation method called S＿CEEMDAN based on the complete ensemble empirical mode decomposition with adaptive noise(CEEMDAN)algorithm is proposed to adaptively separate the second harmonic components for ultrasound tissue harmonic imaging.First,the ensemble size of the CEEMDAN algorithm is calculated adaptively according to the standard deviation of the added white noise.A set of intrinsic mode functions(IMFs)is then obtained by the CEEMDAN algorithm from the ultrasonic RF echo signals.According to the IMF spectra,the IMFs that contain both fundamental and harmonic components are further decomposed.The separation process is performed until all the obtained IMFs have been divided into either fundamental or harmonic categories.Finally,the fundamental and harmonic RF echo signals are obtained from the accumulations of signals on basis of these two categories respectively.In simulation experiments based on CREANUIS,the S＿CEEMDAN-based results are similar to the filter-based results,but better than the pulse inversion-based results.In clinical trials,the ultrasonic RF echo signals of 10 healthy men carotid arteries were collected for the harmonic separation.The results show that: S＿CEEMDAN overcomes the inflexible parameter configuration of the filter method and the motion artifact of the pulse inversion method.It obtains better harmonic B-mode images,which proves the effectiveness of the S＿CEEMDAN method proposed in this paper.2.In order to further improve the efficiency of the adaptive separation method,especially the separation efficiency when one frame of signal contains multiple scan lines and one scan line contains many pixels,an adaptive separation method based on empirical wavelet transform(AEWT)is proposed to quickly separate the harmonics from contrastenhanced ultrasonic RF echo signals.In this method,the spectrum for each RF scan line is calculated by using the Fourier transform,before being preprocessed by removing the spectral tendencies and reducing spectral noise.A meaningful partition based on the scalespace representation is performed on the preprocessed spectrum,from which harmonic boundaries are automatically determined.Thereafter,the multi-order harmonics are obtained using the empirical scaling function and empirical wavelets.Multi-order harmonic separation is achieved after separating all RF scan lines in a frame.In the experiments,the ultrasound RF echo signals collected from the iliac arteries of 5 rabbits before and 10 s,20 s,and 25 s after the injection of the contrast agent are separated using the AEWT method,and the separation results are compared with the results of the 4th-,8th-and 12th-order Butterworth filters.Results show that the AEWT-based harmonic separation can adaptively match different harmonic spectrum changes to obtain more accurate harmonic components,thereby improving the quality of each harmonic image.The AEWT method separates one frame image containing 256 scan lines in 2.63 ± 0.37 s,which is the same order of magnitude as 1.05 ± 0.23 s for the filter-based separation.3.To solve the problem that using original ultrasonic echo signal imaging which is not sensitive enough can’t detect thermal damage in microwave ablation,a harmonic Nakagami parameter imaging method based on the above two adaptive nonlinear harmonic separation is proposed to improve the monitoring performance of microwave ablation.Firstly,S＿CEEMDAN and efficient AEWT adaptive method proposed in this paper are respectively used to separate the nonlinear harmonic components from ultrasonic echo signals and generate corresponding harmonic B-mode images.Then,to improve the accuracy of microwave ablation thermal damage monitoring,the harmonic ultrasound envelope signals are extracted,and the window composite Nakagami parameter imaging and the Gaussian approximation process are used to estimate the thermal damage area.In the experiments,the ultrasonic echo RF signals of 10 groups of pig livers at different times of microwave ablation were collected.The two adaptive separation methods proposed in this paper are used to obtain harmonic B-mode images and Nakagami parameter images by Gaussian approximation.The results are compared with the fundamental and harmonic images obtained by filters.Results show that sensitivity and accuracy of the harmonic images are better than those of the fundamental images.The effect of harmonic images based on the AEWT method are the best,while the images baesd on filter method are the worst.In summary,the S＿CEEMDAN method provides better separation performance owing to its good adaptivity and lower motion artifacts,which makes it a potential alternative to the current methods for harmonic imaging.The AEWT method uses the rigorous theory of wavelet to improve the processing speed of the adaptive algorithm,and optimally track the changes of the signal spectrum,effectively realized the separation of the multi-order harmonics in the ultrasonic echo signal.It improves the contrast and CTR of multi-order harmonic images,provides more accurate diagnostic information for the clinic.In addition,the harmonic Nakagami parameter images by Gaussian approximation can improve the accuracy of ablation thermal damage monitoring,and provide more effective monitoring for the clinical microwave ablation procedure.