Cognitive load in tele-robotic surgery: a comparison of eye tracker designs

Area of Science:

• Robotics
• Human-Computer Interaction
• Ophthalmology

Background:

• Eye gaze tracking and pupillometry are emerging technologies for assessing cognitive load in tele-robotic surgery.
• Current eye and pupil tracking solutions in this domain require further evaluation.

Purpose of the Study:

• To compare the accuracy and stability of three different eye tracker designs (head-mounted and console-mounted) for cognitive load estimation in robotic surgery.
• To assess the suitability of current eye tracking technology for reliable pupillometry signals in tele-robotic surgery.

Main Methods:

• A user study was conducted using the da Vinci Research Kit (dVRK).
• Participants observed markers on the dVRK screen while eye tracking and dVRK video data were collected.
• Pupil detection stability and gaze prediction accuracy were computed and analyzed for three eye tracker designs.

Main Results:

• Head-worn eye trackers demonstrated superior stability and accuracy in gaze prediction and pupil detection compared to console-mounted systems.
• Tracking stability varied across the field of view among the tested trackers.
• Gaze predictions were sometimes detected with high confidence in invalid image zones.

Conclusions:

• Head-worn eye tracking solutions provide enhanced confidence and stability for tele-robotic surgery applications.
• Significant improvements are necessary in pupil detection, tracking stability, and gaze accuracy for eye trackers in tele-robotic scenarios.