Reduced variability of CEUS perfusion estimates in a patient-derived xenograft model via analysis of speckle statistics

Contrast-enhanced ultrasound (CEUS) permits quantification and monitoring of tumor vascular changes in response to anti-angiogenic treatment with the goal of informing targeted therapy. Conventional mean-intensity-based CEUS analysis discounts additional information that may be available from the first-order speckle statistics in a CEUS image. We demonstrate that our compound speckle model for analysis of CEUS images reduces the variability in tumor perfusion quantification, particularly for estimates of blood volume, and reduces the sensitive/resistant classification ambiguity caused by heterogeneous tumor samples in a patient-derived xenograft model of renal cell carcinoma.