Stability analysis of patient-specific 4DCT- and 4DCBCT-based correspondence models

Area of Science:

• Medical Physics
• Radiation Oncology
• Image Guidance

Background:

• Stereotactic body radiation therapy (SBRT) relies on stable patient-specific models linking external breathing signals to internal tumor motion.
• Motion compensation accuracy is critical for effective SBRT delivery.

Purpose of the Study:

• To develop and validate methods for analyzing the stability of patient-specific correspondence models.
• To integrate planning 4D CT and pretreatment 4D CBCT data for stability assessment.
• To correlate model stability with patient-specific clinical parameters.

Main Methods:

• Applied a regression-based approach correlating internal tumor motion (from deformable image registration) with external breathing signals.
• Utilized target volume-based analysis (DSC, ASSD) and system matrix-based analysis (MSD, PCA) to assess correspondence model stability.
• Correlated stability metrics with clinical parameters in 46 lung cancer patients undergoing SBRT.

Main Results:

• Consistent results were observed between the two complementary analysis methods.
• Correspondence model stability was not predominant, with many fraction-wise models failing to meet stability thresholds.
• Model stability did not degrade over the course of treatment; smaller tumor motion ranges correlated with increased stability.

Conclusions:

• The developed methods can identify patients with unstable correspondence models before each fraction.
• These findings serve as indicators for the need for replanning or adaptive treatment strategies.
• Accurate motion modeling is crucial for optimizing SBRT outcomes in lung cancer patients.