Fusion Of B-mode Ultrasound Images Based On The Decibel Peak Difference Of Local Fundamental And Harmonic Spectra

Ultrasonic imaging technology has become one of the important methods for clinical diagnosis,due to the advantage of non-invasion,easy to use,cheap and real-time imaging.The imaging mode of using the echo signal,the ultrasonic transducer received,the frequency the same as the incident sound wave,was called fundamental imaging.It uses the ultrasonic linear characteristics,and has clear contour features,while the internal details are scarce.However,the actual biological tissue is not the ideal medium for the existence of linear properties.Because the organization itself has the character of attenuating and absorbing ultrasonic energy,the acoustic nonlinearity will occur during transmission,accompanying with high order harmonic frequency,which has multiple relationship with the incident sound wave frequency.The imaging method using the high order harmonic echo signal is called harmonic imaging.The image is rich in the details of the organization,b ut vague in features of contour.Therefore,it is difficult to obtain high quality B-mode images with clear contour features and rich internal details using a single imaging mode.It is of great significance to obtain more accurate diagnostic information,if making full use of the complementary advantage information of the fundamental and harmonic images to improve the clarity and expression of the original B-mode image.In this paper,a new method,for the fusion of fundamental and harmonic images,based on fundamental and harmonic spectra decibel peak difference value,is proposed.Firstly,the fundamental and harmonic component was separated from the original ultrasonic radio frequency echo signal,and calculated the spectrum decibel peak difference.The weight of fundamental and harmonic radio frequency signal is determined,based on above.The radio frequency echo signal was fused by the fundamental and harmonic radio frequency signal using the weighted average method.Finally,ultrasound fusion image was generated thought the Hilbert transformation,logarithmic compression,sampling interpolation/drop.The 30 groups inhomogeneous cyst organization model simulation experimental results show that the contrast of fusion images of this method,the weighted average method and wavelet transform method increased by 32%,8% and 8% compared with the original image,respectively.The simulation result was confirmed by the experimental results of the human carotid artery.The contrast of the fusion image of above methods was 30%,7% and 10% higher than the original image in clinical experiments.To sum up,this novel method can realize the complementary advantages of fundamental and harmonic images,highlight harmonic effectiveness caused by the organization nonlinearity.So the high quality B-mode images concluding clear contour characteristics and rich internal details,was obtained.The results will better obtain more accurate diagnostic information.