Three-dimensional estimation of ultrasound-contrast-agent dispersion and convection in the prostate

Prostate cancer (PCa) diagnosis still relies on systematic biopsy due to the insufficient reliability of imaging techniques. In this work, we introduce a new method to quantify contrast agent convective dispersion (D) and velocity (v) in a three-dimensional dynamic contrast-enhanced ultrasound (3D DCE-US) recording. First, the concentration gradients of the data are established using Gaussian derivatives in space and time. Assuming D and v to be locally constant, the convective-dispersion equation can then be solved by minimizing a leastsquares problem within a moving 3D kernel. Six PCa patients underwent two-minute 3D DCE-US examination prior to radical prostatectomy. Subsequently, the 3D parametric maps of D and v were compared with the 3D histopathologic data. Preliminary results suggest a good correlation between malignancy according to histopathology and both D and v maps.