Haptic Feedback in Master-Slave Robotic Laparoscopic Surgery

Haptic Feedback in Master-Slave Robotic Laparoscopic Surgery

de Goede, Nick (TU Delft Mechanical, Maritime and Materials Engineering; TU Delft Biomechanical Engineering)

Horeman, Tim (mentor)
Dankelman, Jenny (graduation committee)
Abbink, David (graduation committee)

Delft University of Technology

2017-08-14

Commercially available laparoscopic surgical robots are not currently equipped with force feedback. This may lead to unnecessary tissue damage when gripping tissue in a laparascopic procedure, as shown in a prior literature study. The goal of this MSc thesis project is the implementation of haptic feedback in a prototype of a master-slave surgical system and a new laparoscopic gripping instrument that is currently under development. Both the master-slave system and the gripping instrument are collaborations between the TU Delft and the Leidse Instrumentmakers School. The master-slave setup is meant as a testing and demonstrating platform for robotic laparoscopic instrument prototypes. It uses the same instrument interface as the da Vinci surgical system (Intuitive Surgical Inc.). The Shaft-Actuated-Tip-Articulation-4-Flexors (SATA-4F) laparoscopic gripper prototype is meant as an alternative to the gripping instruments the da Vinci uses.

The SATA-4F instrument is outfitted with force sensors and the master-slave system is equipped with sensors and actuators.
Different low-end force sensors are compared in terms of linearity, precision, hysteresis, drift and signal-to-noise ratio. A sensor is selected based on the results. Four concepts for implementing the sensors into the SATA-4F instrument are proposed and assessed. The selected concept is implemented. Three concepts for the master handle are also proposed and assessed. After the first prototype failed, a second one is produced. The second prototype has proven to function properly.
System Identification techniques are used to analyze the system components. No models with a target score of 95% are found using the System Identification Toolbox. Three different control strategies are compared and one of them is selected. The Force-Reflection controller is programmed and the system is tested and evaluated.

A simple force-pulling task was designed, where subjects held synthetic tissue with the master-slave robot. A small DC motor mounted on a force measuring platform pulled at the tissue. The participants were tasked with holding the tissue in place, while minimizing grasping forces. Letting novices perform this task results in significantly lower mean maximum gripping forces when force feedback is switched on. However, the participants let go of the tissue more often as well. There were no visible significant learning effects. Reaction forces measured by a platform have no effect on the mean maximum gripping forces. There is also no visible relation between the failed trials and the reaction forces of these trials.

While writing this thesis and working on the project, other students have been working on the project as well. New designs have been made for both the mechanical parts of the master and the slave device. The SATA-4F instrument itself will also be redesigned. The goals presented in this thesis have largely been achieved. However, the master and slave devices need more work. While the mechanical parts of the master and the slave are being redesigned, the control loop and sampling frequency is too low. The 3D motion of both the master and the slave have been disabled because of this. While MATLAB and the LabJack have been extremely useful for designing and testing, it seems that MATLAB and a LabJack are not suitable for the control of a surgical robot.

The setup can now be used for testing and demonstrating prototype laparoscopic instruments. With the new master and slave components and significant other improvements, the setup can be used to compare different haptic feedback controllers as well. It can also be used to do more research in the field of robot-assisted surgery with respect to haptic feedback. If two robotic laparoscopic instruments with all degrees of freedom would be available, typical experiments such as knot-tying or suturing could be performed with and without force feedback to fully investigate the usefulness of the feedback. Of course, force feedback can be implemented in more, if not all degrees of freedom; aside from gripper forces, reaction forces and torques on the instrument itself can also be measured and fed back to a surgeon. This would allow for even more research in the field of robotic surgery with respect to feedback.

laparoscopic surgery
Telemanipulation
Robotics
haptic

http://resolver.tudelft.nl/uuid:aae0240e-9019-4501-9e8e-560c17f04ed6

Student theses

master thesis

© 2017 Nick de Goede