High Frame Rate 3-D Super Resolution Ultrasound Microvascular Imaging

摘要

ABSTRACT Recently, the revolutionary technology of super localization microscopy in the optical imaging domain has been translated into the medical ultrasound domain. By localizing the centers of scattering contrast agents, a similar ultrasound localization microscopy technique, also referred to as contrast enhanced super-resolution (CESR) imaging, has been demonstrated with ultrasound. This novel technique enables imaging of microvessels at resolutions as small as ten micrometers, over an order of magnitude smaller than the ultrasound diffraction limit, and at depths much greater than traditionally limited by frequency. In order to achieve advances in all three of these seeming paradoxical dimensions - super-resolution contrast imaging requires that thousands of frames of data to be analyzed, making this technique much slower than standard ultrasound imaging. The result is that super-resolution imaging would be difficult if not impossible to translate to the clinic in its current form with current clinical hardware, especially if 3-D imaging is desired (which it is for microvascular morphological analysis), as a single 3-D image volume would take tens of minutes to acquire. However, there is a solution to this, which our group proposes to achieve in this project. Recent advances in ultrasound hardware have enabled ultra-high frame rate processing. Our academic and clinical teams at UNC Chapel Hill are partnering with Verasonics, Inc, a world leading industrial partner in next-generation ultrasound systems, to develop and translate the first high-frame rate 3-D super resolution imaging modality to the clinic. We will do this by first designing and constructing a 1024 channel ultra-high frame rate ultrasound system, designed to operate with a 32x32 matrix transducer. Ultra-fast processors, large RAM buffers, GPUs, and high- bandwidth data transfer hardware will be utilized to handle the massive data acquisition and processing. New software and implementation approaches designed at UNC, including our innovative adaptive multi-focus beamforming approach, will further increase sensitivity and resolution at clinically relevant depths, and enable full 3-D volume acquisitions at volume frame rates over 5000 FPS, suitable for fast 3-D super-resolution imaging in humans. Our approach will be validated in phantoms, rodent models of human disease, and in two different clinical applications where ultrasound specificity is limited, breast, and thyroid. Our motivation is to develop super-resolution imaging as a novel new approach for imaging angiogenesis – one of the hallmarks of cancer, as a new biomarker target for both diagnostics and assessment of response to therapy. The ability to differentiate lesions based on microvascular fingerprint, rather than tumor anatomy, would be a paradigm shift in ultrasound diagnostics, and will improve the specificity of ultrasound to malignancy, and advance clinically needed in breast, prostate, thyroid, and other oncological applications. However, the advancement of the proposed technology will undoubtedly open doors to other clinical applications as well, such as wound healing and vasa vasorum imaging in atherosclerosis.