KMID : 1149020210230010047
|
|
Journal of Korean Society of Computed Tomographic Technology 2021 Volume.23 No. 1 p.47 ~ p.56
|
|
Advanced intelligent Clear-IQ Engine (AiCE) of Coronary CT Angiography in 640 MSCT: Effect on Image Quality Compared with Deep Learning Reconstruction and Iterative Reconstruction
|
|
Jeon Sang-Hyun
An Gi-Yong Shim Hak-Joon Ko Sung-Min Jeon Pil-Hyeon
|
|
Abstract
|
|
|
Purpose: The development of Deep Learinig Reconstruction(DLR) technology for CT has various advantages, including reducing exposure dose received by patients, reducing noise in images, and increasing resolution at the same dose. In this work, we wanted to evaluate the image quality of Advanced Intelligent Clear-IQ Engine(AiCE) on clinical utility compared to Filtered Back Projection(FBP) and Iterative Reconstruction(Adaptive Iterative Dose Reduction 3D; AIDR3D) in Coronary CT Angiography.
Material and method: This study was divided into three groups according to BMI, and the subject study was divided into two groups: patient study and phantom study. We reconstruct the images using three algorithms: FBP, AIDR3D, and AiCE in the same patient. Phantom study was designed to represent Aortic root and Left main coronary arty in Lungman Phantom, and reconstructed images using three algorithms. To objectively evaluate image quality, we analyze four parameters: Image noise, CT density, Signal-to-noise ratio(SNR) and Contrast-to-noise ratio(CNR) for each data.
Result: AiCE reduced noise by 51.46 % compared to FBP. It was found that it was reduced by 30.13 % compared to AIDR3D. In addition, regardless of the BMI category, AiCE has a significant noise improvement effect compared to AIDR3D. AiCE showed significantly reduced noise and better image quality in the larger BMI category compared to FBP and AIDR3D.
Conclusion: Compared to FBP, AIDR3D images, we find that AiCE is very effective in reducing image noise and improving parameters of images such as SNR and CNR. It was shown that the improvement effect of AiCE increased with a larger BMI.
|
|
KEYWORD
|
|
Deep learning reconstruction, AiCE(advanced intelligent clear-IQ engine), Coronary CT Agiography(CCTA), Image quality
|
|
FullTexts / Linksout information
|
|
|
|
Listed journal information
|
|
|
|