Yen-Yi Juo, MD, MPH1, Ahmad Abiri2, Anna Tao2, Jake Pensa2, Christina Hanna3, Syed J Askari2, Warren Grundfest, MD4, Erik Dutson, MD1. 1Department of Surgery, University of California, Los Angeles, 2Department of Bioengineering, University of California, Los Angeles, 3Center for Advanced Surgical and Interventional Technology, University of California, Los Angeles, 4Department of Bioengineering, Electrical Engineering, and Surgery, University of California, Los Angeles
Introduction: We aim to carry out a feasibility study testing the use of haptic feedback in robotic surgical systems in discerning anatomical structures of different firmness. Surgeons frequently rely on tactile feedback during laparotomy for localization of critical vascular or neural structures imbedded within soft tissue. We hypothesize that this capability can be simulated with haptic feedback on a surgical robot.
Methods: A haptic feedback system compatible with da Vinci surgical robotic system was developed transmitting forces from piezoresistive sensors mounted on robotic instrument tips to surgeon’s fingertips via balloon actuators. A phantom was fabricated by burying a plastic tubing within a rectangular sponge without causing significant surface deformity. 19 subjects of variable robotic experience were asked to localize the hidden tubular structure using the da Vinci surgical robotic system under three conditions, 1) without haptic feedback, 2) with normal force haptic feedback using pneumatic actuators alone, and 3) with bi-modal haptic feedback, whereby a vibratory component was added and used to warn the user of excessive forces. Repeated-measures ANOVA and McNemar’s test were used to compare task completion time and localization accuracy, respectively.
Results: Mean task completion time and standard errors for trials under the three different feedback conditions were 112.35 +/- 18.62, 97.51 +/- 15.27 and 63.00 +/- 12.11 seconds, respectively. The decrease in task completion time did not reach statistical significance with use of pneumatic normal force feedback alone (p=0.762), but bi-modal haptic feedback use was associated with significantly decreased task completion time (p=0.003). Out of the 19 subjects, 6, 12, and 17 subjects were able to accurately localize the imbedded tubular structure without haptic feedback, with pneumatic normal force feedback alone, and with bi-modal feedback, respectively. Similarly, the increase in discriminatory power was statistically significant only when bi-modal haptic feedback was provided (p=0.0009), but not with pneumatic normal force feedback alone (p=0.058).
Conclusion: Preliminary results from our phantom experiment indicated that haptic feedback in robotic surgery can be helpful with both shortening task time and increasing accuracy in discerning structures hidden in soft tissue. Future studies with larger sample size will help characterize this novel application of haptic feedback in surgical robots.
Presented at the SAGES 2017 Annual Meeting in Houston, TX.
Abstract ID: 80660
Program Number: P431
Presentation Session: Poster (Non CME)
Presentation Type: Poster
