University of Twente Student Theses

Login

Accuracy of head-tracking and eye-tracking when measuring responses to visual cues.

Nassi, T. and Stouten, S. and Wijk, R.J. van and With, A.J.V. de (2016) Accuracy of head-tracking and eye-tracking when measuring responses to visual cues.

[img] PDF
Restricted to Restricted

9MB
Abstract:Introduction: It seems that patients with a hearing implant have a reduced ability to localize sound. A second hearing implant may improve the ability to localize sound, but there is a lack of high level evidence. The Donders Institute for Brain, Cognition and Behaviour investigates whether a reduced ability to localize sound is associated with non-use of hearing implants. Using head-tracking and eye-tracking systems it may be possible to analyze sound localization responses of patients. The aim of this study is therefore to determine the accuracy of head-tracking and eye-tracking. Method: To determine the accuracy of the head- and eye-tracking systems 17 subjects were measured in a standard set-up. The subjects were instructed to follow a specific pattern with their head or eyes, which was recorded with the ART Smarttrack head-tracking system or with the SMI ETG 2 eye-tracker. Using Matlab the obtained data is processed. The accuracy is then determined with SPSS based on validity, reproducibility and repeatability. Results: It is possible to measure head and eye movements using both tracking systems. For head-tracking the mean measured angle of most of the fixation points differs from the desired angle (P \textless 0.05). However, with an acceptance of one degree the validity and reproducibility is acceptable for the less than half of the fixation points. For the repeatability an Intraclass Correlation Coefficient of 1 is found for azimuth as well as elevation. For eye-tracking the mean measured angle of a few fixation points differs from the desired angle (P < 0.05). With the acceptance of one degree the validity and reproducibility is acceptable for most fixation point. For the repeatability an Intraclass Correlation Coefficient of 0.994 and 0.890 is found respectively for azimuth and elevation. Conclusion: Overall it can be concluded that both the head- and eye-tracking systems are unable to determine the exact angle of the point the subject was moving to. Especially the measurement of greater angles tend to show a larger offset then the lower angles. But, the repeatability and thus the ability to distinguish different angles looks promising. In future studies a better room calibration for the head-tracking and a better filtering for the eye-tracking has to be taken in consideration.
Item Type:Essay (Bachelor)
Faculty:TNW: Science and Technology
Subject:44 medicine
Programme:Technical Medicine BSc (50033)
Link to this item:http://purl.utwente.nl/essays/69867
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page