Is a Depth Camera in Agreement with an Electromagnetic Tracking Device when Measuring Head Position?

Area of Science:

• Biomedical Engineering
• Clinical Measurement
• Human Biomechanics

Background:

• Abnormal head postures (AHPs) are clinically observed and measured.
• Depth cameras with face-tracking offer potential for reliable head position measurement.
• This study validates a depth camera against a gold standard system.

Purpose of the Study:

• To compare the accuracy of a depth camera (Microsoft Kinect) with an electromagnetic tracking system (Polhemus) for measuring head position.
• To assess the reliability of depth camera measurements for various head movements.

Main Methods:

• Twenty healthy volunteers participated.
• Head position was simultaneously recorded using Kinect and Polhemus devices.
• Participants performed standardized head movements (chin up/down, turn, tilt).

Main Results:

• Kinect measurements were consistently smaller than Polhemus.
• Accuracy was comparable for head turns and tilts (P > 0.05).
• Significant differences were found for chin up/down movements (P < 0.01), with Kinect being less accurate and more variable.

Conclusions:

• Depth cameras (Kinect) are comparable to electromagnetic systems for head turns and tilts.
• Kinect demonstrates lower accuracy and higher variability for vertical head movements (chin up/down).
• Further research is required before clinical adoption of depth cameras for head position assessment.