Relative Blood Volume Estimation from Clinical Super-Resolution US Imaging in Breast Cancer

In [1] we demonstrated the feasibility of applying super-resolution US imaging in a clinical set-up. However, pronounced in-plane and out-of-plane motion and a non-optimized microbubble injection scheme made the processing more complex than in preclinical investigations and reduced the length of evaluable sequences for the estimation of the clinically relevant parameter relative blood volume (rBV). Here, we apply the saturation model proposed in [2] to the clinical data. This model assumes, that the reconstruction of new vessels saturates to the full vessel tree over measurement time. We compare the results of state-of-the-art contrast-enhanced to super-resolution US imaging and derive objective criteria for the assessment of measurements. Though between 180 s and 350 s would have been necessary to reconstruct 90% of the microvasculature, we found that a reliable prediction of rBV would already have been possible after 20 s to 75 s by applying the model. These times could be reduced further with an optimized MB injection scheme.