Technical Note: Solving the "Chinese postman problem" for effective contour deformation.

Area of Science:

• Medical Imaging
• Radiotherapy
• Computational Anatomy

Background:

• Accurate contour deformation is crucial for image-guided radiotherapy (IGRT).
• Deformable image registration (DIR) is often used for atlas-based segmentation and contour propagation.
• Existing methods may struggle with complex deformations.

Purpose of the Study:

• To develop a practical and accurate method for contour deformation in IGRT.
• To improve contour propagation accuracy using deformable image registration.

Main Methods:

• Developed a 3D mesh-based contour deformation approach.
• Converted 2D contours to 3D triangular meshes, deformed them using DIR vectors.
• Reorganized deformed mesh points using the Chinese Postman Problem (CPP) algorithm.
• Validated against an image-based contour deformation method.

Main Results:

• The 3D mesh-based approach effectively reorganized complex contours.
• Achieved high similarity (Dice: 97.6% lung, 97.5% heart) and low distance (Hausdorff: 19.8mm lung, 6.1mm heart) compared to image-based methods.
• The mesh-based method preserved fine details, particularly for lung contours, unlike the image-based approach.

Conclusions:

• A practical and accurate contour deformation approach using 3D mesh operations was developed.
• The method demonstrates effectiveness for both clinical and research applications in IGRT.
• This technique enhances the reliability of contour propagation in image-guided treatments.