Lung surface deformation prediction from spirometry measurement and chest wall surface motion

Area of Science:

• Medical Physics
• Biomedical Engineering
• Computational Biology

Background:

• Accurate lung motion prediction is crucial for effective radiation therapy.
• Current methods often rely on invasive or complex imaging techniques.

Purpose of the Study:

• To develop and evaluate a noninvasive method for predicting lung surface motion.
• To utilize spirometry and chest/abdomen motion data for this prediction.

Main Methods:

• A patient-specific 3D lung surface mesh was created.
• Spirometry and 192 chest/abdomen surface points were used.
• Canonical ridge regression and QR factorization were employed to build a predictive model.

Main Results:

• The model achieved an average estimation error of less than 1 mm for lung surface vertices.
• Lower lung regions showed slightly higher estimation errors.
• Predicted surface deformation vector fields closely matched real ones.

Conclusions:

• The developed predictive model offers a noninvasive approach to determine lung boundary conditions.
• This system can enhance personalized biomechanical respiration modeling for radiation therapy.