Real-Time External Respiratory Motion Measuring Technique Using an RGB-D Camera and Principal Component Analysis

Area of Science:

• Medical Physics
• Biomedical Engineering
• Radiotherapy Technology

Background:

• Accurate tracking of thoracic and abdominal respiratory motion is critical in external beam radiotherapy to prevent errors in target delineation and minimize healthy tissue radiation exposure.
• Existing respiratory motion modeling techniques often rely on radiography, wearable markers, or implanted nodes, which have limitations.
• Inaccurate respiratory motion tracking can lead to suboptimal radiotherapy outcomes, including increased radiation dose to healthy tissues.

Purpose of the Study:

• To propose and validate a novel, non-invasive respiratory motion tracking approach for radiotherapy using an RGB-D camera.
• To develop a patient-specific respiratory motion model for real-time external motion measurement.
• To enhance tracking accuracy and robustness by incorporating a marker-based depth frame registration technique.

Main Methods:

• Utilized an RGB-D camera to capture depth data for respiratory motion tracking.
• Developed a patient-specific respiratory motion model via principal component analysis (PCA) to denoise depth data.
• Implemented a marker-based depth frame registration technique to define a consistent anatomical measurement region and manage patient movement.
• Validated the system's accuracy against spirometer and laser line scanning measurements.

Main Results:

• Achieved a high correlation coefficient of 0.97 when compared to spirometer data.
• Demonstrated a low average error of 0.53 mm relative to laser line scanning ground truth.
• Successfully created a patient-specific respiratory motion model for real-time external motion measurement.
• The marker-based registration effectively handled patient movements during simulated treatment conditions.

Conclusions:

• The proposed RGB-D camera-based system offers an accurate and non-invasive method for external respiratory motion tracking in radiotherapy.
• This approach has the potential to improve radiotherapy precision by providing reliable motion data.
• Future work will focus on correlating external motion measurements with internal tumor motion for enhanced treatment planning.