Feasibility investigation of logarithmic Nakagami parametric imaging in recovering underestimated perfusion metrics of DCEUS in the uneven acoustic field

Abstract

Owing to acoustic-pressure dependence, amplitudes of backscattered-echoes of encapsulated microbubbles (MBs) are unavoidably regulated by an uneven acoustic field, resulting in the misestimation of hemodynamics in conventional amplitude-coding dynamic contrast-enhanced ultrasound (DCEUS) with focused pulse transmission. This study aimed to investigate the feasibility and performance of Nakagami statistical-feature parametric imaging to recover the above misestimation. Methods: Logarithmic Nakagami parameter (m)-coding DCEUS scheme was investigated via simulation and in vitro MB phantoms as well as in vivo kidneyperfusion experiments of four rabbits in the uneven acoustic fields with two different focal depths. In vivo tissue artifacts for m estimation were suppressed by pulse-inversion second-harmonic imaging and its robustness was enhanced by multiscale moment-estimation strategy. Time-Nakagami-m curves and the corresponding perfusion metrics of intensity and volume were calculated from the logarithmic m-coding DCEUS images within the prefocal and focal regions. These curves and metrics were further compared with the perfusion curves and metrics estimated from the conventional amplitude-coding images within the same regions. Results: Compared with amplitudes of nonlinear scattering MB echoes, their logarithmic m values were relatively independent of the changes in acoustics pressures. Compared with the fixed-scale moment-estimation, the perfusion intensity estimated from logarithmic m-coding DCEUS scheme using multiscale statistical moment-estimation had smaller differences between the prefocal and focal regions. The differences of perfusion intensity induced by an uneven acoustic field decreased to 3.47% ± 1.58 %. The differences decreased by the logarithmic m-coding DCEUS scheme were further regulated by threshold values of m estimation. Conclusions:The logarithmic m-coding DCEUS scheme could recover the underestimated MB backscattered-echoes and the misestimated perfusion intensity induced by the uneven acoustic field. The scheme had the potential to weaken the limitation of microvasculature identification and hemodynamic characterization marked by MBs within tissues or tumors in the uneven acoustic field. K E Y W O R D Scontrast-enhanced ultrasound, focused acoustic field, hemodynamic, Nakagami parameter 2452

Ultrasound Entropy Imaging Based on the Kernel Density Estimation: A New Approach to Hepatic Steatosis Characterization

“…These techniques can be classified into model-based and non-model-based ones. Typical model-based techniques are ultrasound Nakagami imaging [10][11][12][13] and homodyned-K imaging [14,15] that are based on the generalized statistical models of backscatter envelopes. Ultrasound entropy imaging uses the informationtheoretic entropy of ultrasound backscatter envelopes, without a perquisite for satisfying a specific statistical model [9], which is of growing research interest in the field of quantitative ultrasound.…”