Three-dimensional velocity mapping of lung motion using vessel bifurcation pattern matching

Area of Science:

• Medical Imaging
• Radiotherapy Physics
• Computational Anatomy

Background:

• Accurate quantification of three-dimensional (3D) lung motion is crucial for effective radiation therapy.
• Existing methods may require fiducial markers, limiting their applicability.
• Understanding organ deformation is essential for precise dose delivery.

Purpose of the Study:

• To develop and validate a new technique for quantifying 3D lung motion using 4D-CT images.
• To track anatomical features within the lung to derive 3D displacement vectors.
• To assess the accuracy of the technique for potential use in radiation therapy.

Main Methods:

• Utilized sequential 3D-CT (4D-CT) images to model lung volume and extract vessel structures.
• Derived feature points from vessel bifurcations and tracked them across image series.
• Employed point pattern matching with probabilistic relaxation for feature point tracking.
• Validated the method using a lung CT image, an artificially deformed image, and a phantom.

Main Results:

• The developed technique accurately quantifies 3D lung motion by tracking internal anatomical features.
• Achieved vector errors within approximately 1 voxel (1 mm), indicating high precision.
• Demonstrated the ability to quantify realistic 3D organ motion without fiducial markers.

Conclusions:

• The proposed quantification technique offers high accuracy for 3D lung motion, suitable for radiation therapy applications.
• This method allows for the quantification of tumor deformation through gridding interpolation.
• Potential applications include improved dose estimation in mobile organs and 4D treatment planning.