Improving Image Registration in Radiation Therapy
Author Information
Author(s): Yan Hui Lei, Ren Wu, Jackie Di Fu, Yin Fang-Fang
Primary Institution: Duke University Medical Center
Hypothesis
The study investigates the effect of different anatomical information combinations on the accuracy and robustness of image registration algorithms in radiation therapy.
Conclusion
The study found that the combination of Bone + Tissue + Air (BTA) provided the best registration accuracy and robustness.
Supporting Evidence
- The BTA combination achieved the widest attraction range of 73.3 mm and 81.6 degrees.
- The BT combination showed the second largest attraction range.
- The B combination had the smallest attraction range, indicating limited convergence properties.
Takeaway
This study looked at how to make sure images used in cancer treatment are lined up correctly, and found that using more types of body tissue in the images helps a lot.
Methodology
An intensity-based six-degree image registration algorithm was developed and tested using cone-beam CT and planning CT images of an anthropomorphic chest phantom.
Limitations
The results may vary with different similarity measurements, regions of interest, and image enhancement methods used.
Want to read the original?
Access the complete publication on the publisher's website