|
KMID : 1149020180200020079
|
|
Journal of Korean Society of Computed Tomographic Technology
2018 Volume.20 No. 2 p.79 ~ p.87
|
|
|
Study on the usefulness of Metal artifact reduction algorithm and MonoE reconstruction (Dual Layer Detector) overlapping application
|
|
Baek Jang-Hun
Kang Seon-Young
Lee Jun-Hyub
Lee Yun-Sang
|
|
|
Abstract
|
|
|
Purpose: Implant insertion In order to increase the possibility of finding a lesion in the soft tissue of a patient, I tried to examine the usefulness of overlapping application images of MAR algorithm and MoneE reconstruction (Dual Layer Detector).
Material and methods: In order to reproduce the low contrast lesion on the pig forepaw, 2 cc sylinge was mixed with saline and a contrast agent to create two, and a surgical titanum screw was inserted to make Metal artifact. First, the similarity of HU values of Low contrast lesions was evaluated. If Metal artifact is not available, set the ROI for the two lesions, use the obtained HU value as the control group, and when the metal artifact occurs iDose, iDose + O - MAR, IMR, IMR + O - MAR, MonoE + O - MAR the HU values measured by algorithms image applying the set to experimental groups. We examined the conditions under which the experimental groups of two lesions influenced by direct and indirect approach the values of the control group in each algorithm. Secondly, the visibility of low contrast lesions was evaluated. Five radiologists scored a score of ten out of ten and got the average. Third, we compared images with MRI MAR method. For evaluation of MRI and video, 1 cc sylinge diluted the contrast agent and made a phantom. Five people from radiation companies got scores with a score of 10 points to obtain an average value of artifact reduction from MRI images and CT images and overall images.
Result: Analysis of the HU value of the sylinge whose Metal artifact is located at the most serious screw end (black streak artifact) in the phantom study revealed that the Attenuation deviation was 66.9%, 25.2%, 64.7%, 24.7%, when O-MAR was applied It turned out that artifact is decreasing. At MonoE Attenuation deviation decreased from 0.1% to 72.9%. When there is no influence of artifact from the same place (control group) The algorithm closest to the value of HU value 67 HU was measured with the 65.5 HU value of MonoE 170 keV. As a result of analyzing the HU value of the lesion located at the white streak artifact, it was measured at Attenuation deviation 8.6%, 0.2%, 8.6%, 0.1%. For MonoE Attenuation deviation was 1% to 5.5%, and it did not show big difference for each unit of energy. Likewise, when there was no effect of artifact at the same position (control group), the EMG was the most elective for the HU value of 59.6 HU, and the algorithm was measured at 60.9 HU value of IMR - OMAR. IIn the evaluation of visibility of Low contrast lesion, MonoE + O - MAR obtained the highest score of visibility in the case of sylinge which is last direct (black streak artifact) to screw. In the case of sylinge which was indirect (white streak artifact) of screw, it was confirmed that sylinge visibility was not good due to the influence of newly generated artifact after applying O-MAR algorithm. In the evaluation comparing images with MRI, in the case of SEMAC which is the MRI metal artifact reduction technique, it certainly provided better images than CT. In the case of CT, we also received appropriate evaluation through application of various algorithms and MoneE reconstruction.
Conclusion: When there is a Metal artifact Through the clinical image of Phantom and Case review, apply MonoE again to the image to which the basic O-MAR is applied the video of the existing MonoE method through O-MAR and SBI data Metal It was confirmed that artifact reduction is more effective. In addition, as shown in clinical images, we were able to confirm that we were able to increase the possibility of finding lesions that could not be found by metal artifact.
|
|
|
KEYWORD
|
|
|
Metal artifact, O-MAR, MonoE, Dual layer
|
|
|
FullTexts / Linksout information
|
|
|
|
|
|
Listed journal information
|
|
|
|
|