Book Description

The lack of force feedback (haptics) in robotic surgery can be considered to be a safety risk leading to accidental tissue damage and puncturing of blood vessels due to excessive forces being applied to tissue and vessels or causing inefficient control over the instruments because of insufficient applied force. This project focuses on providing a satisfactory solution for introducing haptic feedback in robotics-assisted minimally invasive surgical (RAMIS) systems. The research addresses several key issues associated with the incorporation of haptics in a master-slave (teleoperated) robotic environment for minimally invasive surgery (MIS). In this project, we designed a haptics-enabled dual-arm (two masters - two slaves) robotic MIS testbed to investigate and validate various single-arm as well as dual-arm teleoperation scenarios. The most important feature of this setup is the capability of providing haptic feedback in all 7 degrees of freedom (DOF) required for RAMIS (3 translations, 3 rotations and pinch motion of the laparoscopic tool). The setup also enables the evaluation of the effect of replacing haptic feedback by other sensory cues such as visual representation of haptic information (sensory substitution) and the hypothesis that surgical outcomes may be improved by substituting or augmenting haptic feedback by such sensory cues. To provide realistic haptic feedback, it is necessary to measure forces acting at the tip of the laparoscopic instruments in all appropriate directions, as well as when gripping, cutting or palpating tissue. In order to achieve this, we have incorporated two types of laparoscopic instruments in the testbed: A sensorized da Vinci tool, with the capability of measuring grasping forces provided by several strain gauges embedded in the tool shaft, and a customized instrument, the Tactile Sensing Instrument (TSI), which has been developed in our laboratory for soft-tissue palpation in RAMIS. Two surgical scenarios are considered in this project: Tumor localization in soft-tissue palpation, and endoscopic suturing. The first application is to localize tumors embedded in liver and lung tissue through the single-arm master-slave teleoperation system. Since the stiffness of a tumor is higher than that of healthy tissue, it can be distinguished as a hard nodule during remote palpation. Tactile sensing is a method that can be used in RAMIS to localize cancerous tumors prior to performing ablative therapies. However, its performance is highly dependent on the consistency of the exploration force. Using the customized tactile sensing instrument, the pressure distribution over the tissue is captured and provided as a color contour map on a screen. In order to apply the exploration force consistently over the tissue, different force feedback modalities are incorporated with tactile sensing feedback: Direct reflection of force feedback, visual presentation of interaction forces, and a fusion method utilizing an autonomous force control for the exploration force in the palpation direction and direct reflection of the force measured at the location of the tumor to the operator's fingers through the grasper mechanism of the haptic interface. The problem of incorporating haptic feedback in robot-assisted endoscopic suturing is explored as the next telesurgery scenario. The dual-arm teleoperation setup is used for this application. In order to assess the quality of suturing, we divide the suturing task into two phases: stitching and knot tying. Each phase consists of several well-specified sub-tasks. The experiments are performed in three modes: without force feedback, with visual force feedback and with direct force reflection to the user. Three levels are considered for the visual feedback presented to the user. The main objective of showing force in different levels is to assure the user that the force being applied on the suture is sufficient to end up with a secure knot. The main focus on this work is to explore which way of presenting force feedback can be more effectively used, and how each modality can help the user to increase the performance.