An Experimental Comparison Towards Autonomous Camera Navigation to Optimize Training in Robot Assisted Surgery

Abstract

Robot-Assisted Surgery enhances vision and it can restore depth perception, but it introduces the need for learning how to tele-operatively control both the surgical tools and the endoscope. Together with the complexity of selecting the optimal viewpoint to carry out the procedure, this requires distinct training. This work proposes an autonomous camera navigation during the initial stages of training in order to optimize the learning of these skills. A user study involving 26 novice participants was carried out using the master console of the da Vinci Research Kit and a virtual reality training environment. The subjects were randomly divided into two groups: the control group that manually controlled the camera as in the current practice and the experimental group that underwent the autonomous navigation. After training, the time-accuracy metrics of the users who underwent autonomous camera navigation were significantly higher with respect to the control group. Additionally, autonomous camera navigation seemed to be capable to provide an imprinting about endoscope management.