Regularized Spectral Log Difference Technique for Ultrasonic Attenuation Imaging

Abstract

The attenuation coefficient slope (ACS) has the potential to be used for tissue characterization and as a diagnostic ultrasound tool, hence complementing B-mode images. The ACS can be valuable for the estimation of other ultrasound parameters such as the backscatter coefficient. There is a well-known tradeoff between the precision of the estimated ACS values and the data block size used in the spectral-based techniques such as the spectral-log difference (SLD). This tradeoff limits the practical usefulness of the spectral-based attenuation imaging techniques. In this paper, the regularized SLD (RSLD) technique is presented in detail, and evaluated with simulations and experiments with physical phantoms, ex vivo and in vivo. The RSLD technique allowed decreasing estimation variance when using small data block sizes, i.e., fivefold reduction in the standard deviation of percentage error when using data block sizes larger than and more than a tenfold reduction when using data blocks. The precision improvement was obtained without sacrificing estimation accuracy (i.e., estimation bias improved in 70% of the cases by 10% of the ground truth-value on average while degraded in 30% of the cases by 3% of the ground truth-value on average). The improvements in precision allowed for better differentiation of inclusions especially when using small data blocks (i.e., smaller than ) where the contrast-to-noise ratio improved by an order of magnitude on average. The results suggest that the RSLD allows for the reconstruction of attenuation coefficient images with an improved tradeoff between spatial resolution and estimation precision.

Low Variance Estimation of Backscatter Quantitative Ultrasound Parameters Using Dynamic Programming

“…Thus, these parameters can be considered piecewise continuous. [19] This condition can be used to improve parameter estimation. Similar to our previous work in the field of elastography [24], [21], we proposed a regularized cost function that incorporates both data terms and prior information for parameter estimation.…”

“…Although this method substantially improved the results compared to the commonly used Reference Phantom Method (RPM) [17], it calculates these parameters at each spatial location independent of its neighbors and hence neglects spatial dependency of these coefficients. More recently, Coila et al [18], [19] adapted the conventional spectral log difference technique for attenuation estimation by adding a regularization term. Despite substantially improving the results, this work only estimates attenuation (not backscattering).…”

On Multifractal Tissue Characterization in Ultrasound Imaging

“…These works are commonly addressed as tissue characterization or quantitative US. Most of them are based on the estimation of statistical or spectralbased parameters from image regions extracted from different tissues (e.g., [1]) or on the estimation of acoustic parameter maps such as the attenuation coefficient (e.g., [2]).…”

Deconvolution for Improved Multifractal Characterization of Tissues in Ultrasound Imaging

“…Such quantitative measurements, very useful in computer-aided screening tools, are generally extracted from US images such as beamformed radiofrequency (RF), envelope, computed by demodulation of individual RF signal, or B-mode, log-compressed envelope, images. The most used parameters to characterize the tissues rely on acoustical properties (e.g., attenuation, speed of sound, backscattering coefficient [1]) or statistical and spectral information (e.g., [2]). In addition, several studies showed the interest of extracting from US images fractal or multifractal parameters, potentially related to the fractal or multifractal behavior of tissues in space and time (e.g., [3]).…”