Related Experiment Videos
Automated CT marker segmentation for image registration in radionuclide therapy
P Papavasileiou1, G D Flux, M A Flower
1Joint Department of Physics, Institute of Cancer Research and Royal Marsden NHS Trust, Sutton, Surrey, UK. periklis@icr.ac.uk
Physics in Medicine and Biology
|January 5, 2002
Summary
This study introduces an automated CT marker segmentation method for precise image registration. The novel technique achieved accurate 3D localization and registration, with low errors in simulated and patient data.
Area of Science:
- Medical Imaging
- Image Registration
- Computational Anatomy
Background:
- Accurate image registration is crucial for effective radiotherapy and diagnosis.
- Current methods for CT marker segmentation can be time-consuming and prone to error.
Purpose of the Study:
- To develop and validate a novel, automated CT marker segmentation technique for improved image registration.
- To assess the accuracy of the proposed method in both simulated and patient datasets.
Main Methods:
- A novel technique analyzing CT slice contours to identify external markers as contour protrusions.
- Knowledge-based criteria using marker shape and dimensions for identification and segmentation.
- Localization of 3D marker centroids via an intensity-weighted algorithm, followed by least-squares fit registration.
Main Results:
- The automated technique successfully segmented CT markers and localized their centroids.
- Mean residual 3D registration errors were 1.8 +/- 0.3 mm for simulated studies.
- Mean residual 3D registration errors were 4.3 +/- 0.5 mm for patient studies undergoing radionuclide therapy.
Conclusions:
- The developed automated CT marker segmentation technique provides accurate image registration.
- The method shows potential for enhancing precision in medical imaging applications, particularly in radionuclide therapy.
- Further validation in diverse clinical scenarios is warranted to fully establish its utility.