Evaluating commercial image registration packages for radiotherapy treatment planning

Area of Science:

• Medical Imaging
• Radiotherapy Physics
• Oncology

Background:

• Multi-modality imaging is integral to cancer care, from diagnosis to follow-up.
• Reliable image fusion and registration algorithms are essential for accurate treatment planning, particularly in radiotherapy.

Purpose of the Study:

• To assess the accuracy of commercial image fusion software for multi-modality imaging.
• To evaluate the performance of various registration algorithms in the context of radiotherapy target volume delineation.

Main Methods:

• A National Electrical Manufacturers Association (NEMA) body phantom was utilized for evaluation.
• Computer tomography (CT), magnetic resonance (MR), and positron emission tomography (PET) images were fused.
• The study considered fiducial marker selection and window level effects on target size.

Main Results:

• Fusion accuracy for multi-modality images was generally within 0.5-1.5mm of actual feature diameters.
• Volume accuracy was typically less than 2 ml compared to actual values.
• CT image-based fusions demonstrated particular accuracy.

Conclusions:

• The evaluated registration methods provide reliable multi-modality image fusion for oncology applications.
• Accuracy achieved supports the use of these techniques for precise radiotherapy target delineation and treatment planning.