gms | German Medical Science

Introduction: Many studies use eye-tracker to analyse socio-technical systems, also in medical research. Eye-trackers were used as a diagnostic tool for patients, as an input device for people with disabilities or as a tool to analyse cognitive processes in many different domains [1]. Only few articles describe the use of eye-trackers to examine the interaction of human and computer in a critical care environment especially in the field of anaesthesia or surgery [2], [3], [4], [5]. In further studies we want to analyse the socio-technical-system in critical-care-environments to increase patient safety. Therefore, we performed a usability test with head-mounted eye-trackers of three different manufactures in a simulated anaesthesia surrounding. The research questions were:

1. Is the use of eye-tracker in anaesthesia workspaces applicable?

2. Are there any limitations in the use of eye-tracker in these environments?

Eye-tracker must accomplish the following requirements:

1. Rapid focus adaptation and rapid adaptation when brightness changes

2. Good correlation between the real scene and the recorded video frame

3. Proper fixation of the head mounted eye-tracker

Methods: In an anaesthesia workspace we simulated three different scenarios to test and compare the features of the following products: Ergoneers Dikablis Eye-tracking Glasses Professional [6], SMI Eyetracking Glasses 2.0 [7] and Tobii Pro Glasses 2 [8]. We analysed recorded videos with following video software: Avidemux 2.6.18 [9], Tobii Pro Lab 55.5126 and BeGaze 3.6 build 52.

We followed a protocol with detailed description of the tests: (1) Change of gaze during a simulated anaesthesia to test the focus adaptation and the adaptation when brightness changes, (2) Oral intubation to analyse the correlation between the real scene and the recorded video frame and (3) Resuscitation with heart pressure massage to check the fixation of the head mounted eye-tracker.

The eye-trackers were tested by one anaesthesiologist. All recorded videos were analysed independently by two researchers, a computer scientist and a physician (anaesthesiologist).

Result: Every tested eye-tracker had to adapt itself after changing the gaze direction, mostly in less than one second.

The recognition of details on displays depend on the display background and on the display lightning. Very difficult is the recognisability of display with light display background and when room light is switched off. You can see a lot of details on the recorded video of all eye-trackers when looking into the room or into the face of the manikin.

In laryngoscopy there a differences between the real field of vision and the recorded videos in all tested eye-trackers.

Rapid head movements can change the calibration of the eye-tracker.

Conclusion: Possible reasons were detected causing the differences between the eye-trackers: Video formats and different codecs vary [6], [7], [8] furthermore we only used one setting in the eye-tracker-software. Another reason according to the difference between real field of vision and recorded video in laryngoscopy can be that anaesthesiologists do an inclination when they do a laryngoscopy. Using a fixation could attain more stability while head movements.

Die Autoren geben an, dass kein Interessenkonflikt besteht.

Die Autoren geben an, dass ein positives Ethikvotum vorliegt.