Generation of a local lung respiratory motion model using a weighted sparse algorithm and motion prior-based...

Area of Science:

• Medical Imaging
• Computational Anatomy
• Radiology

Background:

• Respiratory motion introduces significant uncertainty in lung tumor localization during percutaneous interventions.
• Accurate estimation of tumor and surrounding vessel motion is critical for effective treatment planning and delivery.

Purpose of the Study:

• To develop and evaluate a novel local motion modeling method for enhanced respiratory motion estimation in lung percutaneous interventions.
• To improve the accuracy of tumor motion prediction in the region of interest (ROI) using adaptive registration techniques.

Main Methods:

• Proposed a weighted sparse statistical modeling (WSSM) method for accurate landmark point error capture and lung motion prediction.
• Developed an adaptive motion prior-based registration method utilizing a B-spline scheme to refine local motion information.
• Evaluated the methods on 15 end-expiratory/end-inspiratory CT image pairs and 31 four-dimensional CT (4DCT) datasets.

Main Results:

• The WSSM method demonstrated superior motion prediction performance compared to existing lung statistical motion modeling approaches.
• The adaptive motion prior-based registration method significantly improved the accuracy of local motion information within the ROI.
• The proposed techniques effectively addressed respiration-induced tumor location uncertainty in lung interventions.

Conclusions:

• The developed WSSM and adaptive registration methods offer a robust solution for accurate respiratory motion estimation in lung percutaneous interventions.
• These advancements have the potential to enhance the precision and safety of lung interventions by providing more reliable tumor motion tracking.