Pericardiocentesis is the drainage of excessive fluid in the pericardium surrounding the heart under ultrasound guidance. The percutaneous procedure comes with the risk of cardiac perforation since the needle is inserted towards the beating heart. This letter presents a novel percutaneous robot that features a flexible wrist made of shape memory alloy (SMA) spring which can vary its stiffness through temperature variation. The SMA spring can be heated and thus stiffened to allow insertion. At lower temperature, it behaves like a spring backbone that allows the needle to be bent away from the heart and thus minimizing the chance of cardiac perforation. A heating tube made of nichrome wire and PTFE tube was developed to be the heat source for the SMA spring. The robot has an outer diameter of 4 mm, similar to the dilator used in pericardiocentesis, and a lumen to pass a guidewire through. A stiffness model and kinematics model were developed and verified through experiments. The surface temperature of the SMA-based wrist was evaluated to be less than the tissue necrosis temperature. Lastly, we demonstrated that the robot could be inserted into gelatin phantoms of different stiffnesses and produce a wristed motion under ultrasound guidance.

This paper presents a novel surrogate-based cross-correlation (SBCC) framework to improve the correlation performance for practical particle image velocimetry (PIV). The basic idea is that an optimized surrogate filter/image, replacing one raw image, will produce a more accurate and robust correlation signal. Specifically, the surrogate image is encouraged to generate a perfect Gaussian-shaped correlation map to tracking particles (PIV image pair) while producing zero responses to image noise (context images). The problem is formulated with an objective function composed of surrogate loss and consistency loss. As a result, the closed-form solution provides a multivariate operator that could consider other negative context images. Compared with state-of-the-art baseline methods (background subtraction, robust phase correlation, etc.), our SBCC method exhibits significant performance improvement (accuracy and robustness) in the synthetic dataset and several challenging experimental PIV cases. Moreover, our implementation with all details (https://github.com/yongleex/SBCC) is also available for interested researchers.