Advertisement

Multi-component video-based feedback using additional webcam inputs improves daVinci surgical skills simulator (DVSSS) performance

Login to Access Video or Poster Abstract: MP51-19
Sources of Funding: none

Introduction

Ergonomics is the study of the efficient movement in working environments. The da Vinci operating system is unique in that it consists of a separate human-machine interface at the operating console including master controls, camera pedal, and clutch. However, current robotic training approaches lack the ability to deliver specific feedback addressing each component. We demonstrate the use of additional web-camera inputs, offering multi-component summative feedback to improve virtual reality (VR) daVinci surgical skills simulator (DVSSS) training.

Methods

18 medical students were enrolled in a study on a VR DaVinic surgical skills curriculum. They were randomized into three groups: Group A (n=6, control), no performance feedback; Group B (n=6, standard formative expert feedback), Group C (n=6, summative expert feedback using multicomponent video feedback of the VR task + webcam feedback of master controls and foot pedal). Each trainee completed each task six times. 4 tasks (Peg Board 2, Camera targeting 2, Ring walk 3 & Suture sponge 3) were chosen. Simulator-measured performance metrics included differences in total score, time and economy of motion over the five trials. Data were analyzed using SPSS version 15.

Results

A learning curve was observed across the five trials in all groups. A significant difference was seen between the three groups for change in overall score across the five trials. Ergonomic metric assessment showed that Groups B and C performed better than Group A (P .002 and P .000, respectively) and that the multicomponent feedback was more effective in tasks involving the use of multiple controls (Camera targeting & Ring walk).

Conclusions

Multi-component summative feedback (combination of task, master control, and camera pedal) is effective in significantly shortening the learning curve in the robotic training process, especially in complex tasks. _x000D_

Funding

none

Authors
Scott Quarrier
Aisha Siebert
Ahmed Ghazi
back to top