Experimental study of the relationship between microbubble size and spatiotemporal pulse sequencing during super-resolution ultrasound imaging

Super-resolution ultrasound imaging (SR-US) has broken the ultrasound (US) diffraction limit and enabled a 10-fold improvement in spatial resolution. Clinical adoption of SR-US is currently limited in part due to long image acquisition times. This study evaluated the use of different-sized microbubbles (MBs) with nonlinear B-mode US and linear contrast-enhanced US (CEUS) imaging strategies. The main endpoint was MB detection rates. Custom US simulations using the Rayleigh-Plesset-Marmottant (RPM) model were compared to experimental US images from a vascular flow phantom. US imaging was performed using a programmable US system (Vantage 256, Verasonics Inc) equipped with an L11-4v linear array transducer and custom pulse sequencing. Four different size-isolated MBs were studied and had diameters that ranged from 1 to 2, 3 to 4, 4 to 5, or 5 to 8 $\mu$ m (Advanced Microbubbles Inc). Both simulation and experimental results revealed that MB size strongly influences CEUS images and contrast-to-tissue ratio (CTR) measurements. As MB size increased beyond 2 $\mu$ m, nonlinear CEUS imaging exhibited a progressive decrease in CTR values, whereas B-mode US showed the opposite trend (range 3. S to 11.2 dB). Overall, our findings reveal that SR-US image quality is considerably impacted by MB size. A combination linear B-mode US and nonlinear CEUS imaging strategy represents one solution to increase the MB detection efficiency and minimize acquisition times required for SR-US image formation. This appears particularly relevant when using polydisperse MB contrast agents like those approved clinically for human studies.