A joint time-frequency approach to mean scatterer spacing estimation

摘要

Real-time ultrasound is an important diagnostic tool for liver disease. For precise diagnosis, quantitative techniques based on ultrasonic tissue characterization are preferred. Mean scatterer spacing is a potential quantitative method for characterizing biological tissue structures, especially for liver tissue. Several techniques were developed for MSS estimation. Two among them are the Wagner's method which is a time domain technique and the Simon's method which is a frequency domain technique. Simon's method is a robust and computationally efficient algorithm. However, for MSS estimation based on real-life data, its performance is limited by the low frequency artifacts caused by the quadratic transformation. For the Wagner's method, the major source of error is the ambiguous correlation peaks due to the speckle fluctuations. Since the estimation errors of the time and frequency domain methods are different when scatterer regularity is low; so based on the estimates of both methods, a joint estimator can be developed to improve the accuracy of MSS estimation. The main objective of this work was to investigate the potential of this new method to estimate MSS when applied to simulated and real backscattered echoes from in vivo liver. In real data experiment of human liver, the joint time and frequency approach to MSS estimation has a better separation for fibrosis stage 1 and stage 3. than Simon's method.