The Use of a Surgery Training Interface (sti) to Guide Optimal Gaze Control Strategies During the Acquisition of a Laparoscopic Skill

S J Vine, PhD, C Lin, PhD, J McGrath, FRCS MD, E Bright, MD, R Masters, DPhil, M R Wilson, PhD. The University of Exeter and The Royal Devon and Exeter Hospital, UK.

Video trainer (VT) and Virtual reality (VR) simulators are becoming increasingly popular tools for the training of laparoscopic surgeons. However, the most effective way to use these tools is still not clear, despite recent developments in theoretically-driven training curricula. Recent research using eye tracking technology has shown that gaze control parameters differentiate experts and novices performing training tasks on VR simulators. As a result researchers have suggested that training novices to adopt an expert’s gaze control may improve the expediency and robustness of the learning of fine motor skills required in laparoscopic surgery.

We have developed the surgery training interface (STI), an evidence based intervention designed to guide the gaze control of trainee surgeons whilst performing training tasks on a VT simulator. The STI software takes the live feed from the camera on a VT box and adapts the image so that only the relevant locations within the scene are visible. By simply highlighting the relevant locations and darkening out the irrelevant areas users are guided to adopt an optimal gaze control strategy which is indicative of an expert.

In order to examine the potential benefits of this approach 13 novices were divided into either a gaze trained (using the STI) or a control group (normal view of the scene). Whilst wearing an ASL mobile eye tracker, participants performed 50 trials of a simple ‘pea’ transfer task followed by a series of retention and transfer tests designed to assess learning and robustness under more demanding conditions. Results show that in a retention test the gaze trained group were significantly faster (p < 0.01) and made fewer errors (knocking the peas off, p < 0.01) than the control group. Participants who were gaze trained also performed significantly better in a transfer task requiring them to perform both the VT task and a tone counting task concurrently (p < 0.05). Finally in a pressure test designed to stress the participants (by setting a time limit), the gaze trained group were significantly faster (p < 0.05) than their control group counterparts. Results corroborate findings from the sport and motor control literature which suggest that adopting expert gaze control strategies can improve the speed of learning and the robustness of performance under pressure or additional attentional load.

Although further work is required to examine the benefits of such an interface, we propose that this would be a useful addition to existing VT trainers. While some VR simulators do guide operators to the areas of importance (by having them flash or change color), their cost may mean that they are beyond the budget of many surgical departments. The STI is therefore a lower cost option that helps guide effective visuo-motor control.

Session: Emerging Technology Poster
Program Number: ETP063
View Poster

« Return to SAGES 2011 abstract archive