Technical Note: Kinect V2 surface filtering during gantry motion for radiotherapy applications

Area of Science:

• Medical Physics
• Biomedical Engineering
• Radiotherapy Technology

Background:

• The Kinect V2 camera shows promise for radiotherapy applications like patient positioning and motion tracking.
• Surface reflections can cause depth measurement errors, impacting accuracy during gantry motion and respiratory signal extraction.

Purpose of the Study:

• To develop and validate filtering techniques for reducing noise in Kinect V2 depth measurements for radiotherapy.
• To improve the accuracy of 1D respiratory signals and 2D surface data derived from the Kinect V2 camera.

Main Methods:

• Implemented a trend removal filter for the 1D respiratory signal.
• Applied a temporal median filter to the 2D surface data.
• Validated filtering techniques using a phantom simulating radiotherapy conditions with ground truth.

Main Results:

• Filtered respiratory signals showed a significantly higher correlation (0.76) compared to unfiltered signals (0.13).
• Surface filtering reduced the mean square distance error by 85% between reference and measured point clouds.

Conclusions:

• The proposed filtering methods effectively compensate for noise and improve surface accuracy.
• Kinect V2 camera, with filtering, shows potential for reliable respiratory tracking and collision detection in radiotherapy.