Date of Original Version



Conference Proceeding

Abstract or Description

Physiological hand tremor and other manual positioning errors limit precision in microsurgical procedures. Our research has involved development of adaptive algorithms and neural network methods for real-time compensation of such errors. This paper presents a novel design for an active hand-held microsurgical instrument to implement these algorithms, particularly during vitreoretinal microsurgery. The basic vitreoretinal instrument consists of a handle fitted with a narrow intraocular shaft to provide access to the interior of the eye. In our design, the instrument handle incorporates six-degree-of-freedom inertial sensing to determine the three-dimensional position of the instrument tip. The intraocular shaft is attached to the instrument handle via a miniature parallel manipulator with three degrees of freedom, controlled by three piezoelectric elements. The manipulator actuates the intraocular shaft in pitch, yaw, and axial extension, allowing the system to perform active compensation of errors in the position of the tip of the intraocular shaft. The paper includes the formulation of the inverse kinematics of the instrument in a manner suitable for online computation. A discussion of practical design considerations and the methods and results of preliminary experiments are also presented.