Abstract
An area of active research involves using the radiation force of ultrasound to expel small kidney stones or fragments from the kidney. The goal of this work is real‐time motion tracking for visual feedback to the user and automated adaptive pushing as the stone moves. Algorithms have been designed to track stone movement during patient respiration but the challenge here is to track the stone motion relative to tissue. A new algorithm was written in MATLAB and implemented on an open‐architecture, software‐based ultrasound system. The algorithm was first trained then implemented in real‐time on B‐mode IQ data recorded from phantom experiments and animal studies. The tracking algorithm uses an ensemble of image processing techniques (2‐D cross‐correlation, phase correlation, and feature‐edge detection) to overlay color on the stone in the real‐time images and to assign a color to indicate the confidence in the identification of the stone. Camera images as well as ultrasound images showed that the system was able to locate a moving stone, re‐target, and apply a new focused push pulse at that location. [Work supported by NIH DK43881, NIH DK086371, and NSBRI through NASA NCC 9‐58].
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call