Affordable sensor based gaze tracking for realistic psychological assessment.

Area of Science:

• Ophthalmology
• Biomedical Engineering
• Computer Science

Background:

• Eye movement analysis is crucial for medical screening and rehabilitation.
• Infrared sensor eye trackers are popular but costly and require frequent recalibration.
• Existing eye trackers suffer from variable and systematic noise, leading to inaccurate gaze tracking.

Purpose of the Study:

• To develop a one-time calibration module for eye trackers.
• To design an algorithm for effectively removing variable and systematic errors in gaze data.
• To improve the accuracy and efficiency of eye gaze tracking.

Main Methods:

• Development of a novel one-time calibration module.
• Implementation of an algorithm to correct gaze tracking data for variable and systematic errors.
• Validation of the approach using Digit Gazing task and standard recall-recognition test.

Main Results:

• The proposed method achieved 90% accuracy in detecting gaze positions for the Digit Gazing task.
• An 82% accuracy was achieved for the standard recall-recognition test.
• Accurate gaze tracking is feasible with a single calibration if the experimental setup remains unchanged.

Conclusions:

• A one-time calibration method can effectively eliminate the need for repeated recalibration in eye trackers.
• The developed algorithm successfully removes noise, enhancing gaze tracking precision.
• This approach offers a cost-effective and accurate solution for eye movement analysis in various applications.