|
KMID : 0390320200300020039
|
|
Chungbuk Medical Journal
2020 Volume.30 No. 2 p.39 ~ p.47
|
|
|
Characteristics of Image Fusion in SyntegraR System
|
|
Kim Won-Dong
|
|
|
Abstract
|
|
|
Purpose : In 2016, Pinnacle 9.10, radiation therapy planning system and the SyntegraR, image fusion system(Philips Medical Systems, The Netherlands) were installed in our hospital. The aim of this study was to describe the characteristics and evaluate the clinical usefulness of the SyntegraR system by comparing three gray-value based automatic image fusion methods (cross correlation[CC], local correlation[LC], and mutual information[MI]).
Materials and Methods : The ‘fusion window’, ‘image fusion algorithm’, ‘viewing windows’, ‘set up the image sets’, and ‘automatic registration’ functions were characterized through work analysis and manual instructions. For the efficiency analysis of fusion method, we fused CT-MRI images from five brain cancer patients and CT-PET images from five lung cancer patients who received radiotherapy and the average time to complete each image fusion was measured.
Results : CC is a useful tool for the fusion of CT-CT and MRI-MRI images obtained with the same image modality and a similar imaging protocol. LC, which applies the CC method in a local sub-region, is suitable for fusing images from different imaging modalities, such as CT-MRI, if they provide the same anatomical information. MI, which assumes that a functional correlation between gray values does not exist, is suitable for fusing images from various different imaging modalities. For the same imaging modality(CT-CT and MRI-MRI fusion), the CC fusion method was the fastest with an average of 18 seconds. The time required for fusing images from different imaging modalities was 27 seconds for LC and 60 seconds for MI in CT-MRI fusion, and 39 seconds for LC and 65 seconds for MI in CT-PET fusion, excluding CC with proven fusion inaccuracy. LC was faster than MI and MI completed the fusion 12 times faster than the manual registration.
Conclusion : The characteristics and efficiency of the SyntegraR image fusion program have been described. This will help to identify the optimal image fusion method to accurately extract the radiation treatment target and eventually contribute to improving the effectiveness of radiation treatment.
|
|
|
KEYWORD
|
|
|
SyntegraR, Image fusion
|
|
|
FullTexts / Linksout information
|
|
|
|
|
|
Listed journal information
|
|
|
|
|