Development of Multisegment Steering Mechanism and 3-D Panorama for Automated Bladder Surveillance System

Abstract

A cystoscope is an invaluable tool for bladder cancer surveillance and lower urinary tract pathology diagnosis. The performance of cystoscopy can be improved by automating the procedure to ensure rapid and complete inspection, having a predicted pathway for imaging. This paper proposes a versatile, wire-driven, multisegment mechanism for steering an imaging probe to follow optimal, preprogrammed scan trajectories. A novel flexible imaging probe scanning fiber endoscope (SFE) of 1.2 mm in diameter generates high-quality reflectance or fluorescent images. A 3-D panoramic reconstruction view of the bladder is generated by custom software to confirm comprehensive surveillance and improve clinical efficiency. The kinematics and controllability of the proposed mechanism are analyzed. The multiple degrees-of-freedom of the mechanism provide the advantage of maintaining the camera perpendicular to the bladder wall with a safety distance. The measured experimental results show 8.1% and 5.1% tip position and orientation errors, respectively, to the theoretical simulation predictions. Within a synthetic bladder phantom made from patient-specific geometry information, the apparatus is able to generate a 3-D panorama of about 60% of the bladder's inner surface, while accepting 50% image overlap between adjacent images. The novel imaging capability of the SFE, combined with a robotized multisegment steering mechanism and an image stitching software, proposes potential advancements to the multimodal 3-D panorama of the bladder.