A Fundamental Theory for Dexterous Surgical Skills Transfer to Medical Robots
Technical Report,15 Sep 2018,14 Sep 2019
Purdue University West Lafayette United States
Pagination or Media Count:
This research focuses on the Medical Robotics Research topic area. Within this area we will address the following objectives 1 mitigate the deleterious effects of signal latency on complex tele-robotic surgical tasks 2 extend to full automation when deemed necessary 3 develop semi-autonomous robotic assistant protocols and 4 develop models for knowledge representation of semi-autonomous medical behaviors. When tele-operation is challenged due to limited bandwidth, latency and lost-of-signal, autonomy needs to step in. While modern robots have been endowed with excellent sensing and dexterous capabilities, there is a gap in knowledge about two critical aspects 1 how to leverage operators expertise under limited connectivity rather than switching onoff autonomy 2 how to transfer automatically existing abundant knowledge about surgical maneuvers from the operating room OR to new, uncontrolled and austere settings. In this research, we will study and theorize about new approaches to predict information to account for randomly delayed and imperfect information due to bandwidth limitations. This will allow us to maximize informational content and minimize redundancy, under constrained communication settings. Based on the predicted information, we will determine at what extent a robot needs to autonomously complete surgical procedures, making use of a sequence of maneuvers associated with the required surgical procedure extracted from a previously learned library. To populate this library, we will theorize new approaches for transfer learning, where learned patterns will be projected to fundamentally different domains with variable resource constraints. This first year the effort were allocated to the creation of a database of multiple robots performing a fundamentals of laparoscopic task peg and pole transfer which was used to create a library of maneuvers and apply principles of transfer learning to recognize such maneuvers.
- Human Factors Engineering and Man Machine Systems