In vivo evaluation of an objective method to select power Doppler wall filter cut-off frequency for microvascular quantification

Operator-dependent settings such as the wall filter cut-off frequency control the incidence of artifacts in Doppler images and limit the reliability of comparisons of Doppler vascular indices across longitudinal studies. To address this problem, the wall filter selection curve (WFSC) method was developed to objectively select the cut-off frequency. This paper reports the first evaluation of the effectiveness of the WFSC method for imaging a tumor model. 3-D power Doppler data were acquired from murine breast cancer tumors over a six-week study and the WFSC method was applied to select a cut-off frequency for the wall filter. Images reconstructed using the WFSC method were compared to images processed using a fixed cut-off frequency. Selection curves for tumor images were consistent with flow-phantom data from previous studies. The mean WFSC-selected cut-off showed significant variation among images acquired at different time points from each tumor (two-way ANOVA, p <; 0.0001), which demonstrates the relevance of adjusting the cut-off frequency to match changing flow conditions. 3-D images processed using the WFSC method depicted additional vascular structures compared to volumes processed using a fixed cut-off frequency. Therefore, the WFSC method can facilitate acquisition of reliable Doppler images by inexperienced or time-sensitive operators.