Predicting respiratory motion for active canceling during percutaneous needle insertion

Abstract

Prediction of bodily motion due to respiration was investigated preparatory to implementation of active compensation for respiration in a robot-assisted system for percutaneous kidney surgery. Data for preliminary testing were recorded from the chest wall of a subject using an optical displacement sensor. The weighted-frequency Fourier linear combiner algorithms an adaptive modeling: algorithm, was used to model and predict respiratory movement. Preliminary results are presented, in which the algorithm is shown to track a 0.86 mm rms respiration signal with 0.11 mm rms error. The general robotic system and compensation scheme are also described.