Effect of AIIR Reconstruction Algorithm on Image Quality of
70 kVp Low Tube Voltage Scanned Aortic CTA
YOU Yongchun1,LI Wanjiang1,ZHONG Sihua2,SHUAI Tao1,LIU Hongchuan1,LI Zhenlin1
1. Department of Radiology, West China Hospital, Sichuan University, Chengdu Sichuan 610041, China丨
2. Central Research Institute, United Imaging Healthcare, Shanghai 201870, China
Abstract:Objective To explore the best reconstruction level of artificial intelligence iterative reconstruction in 70 kVp low tube
voltage aortic computed tomography angiography (CTA). Methods A total of 50 patients who underwent aortic CTA scan in our
hospital from March to May 2021 were prospectively collected. The original data were used to reconstruct the images of Karl 5
and AIIR 1-5, and 6 groups of reconstructed images were obtained. The CT and SD values of ascending aorta, descending aorta,
abdominal aorta, left iliac artery and right iliac artery were measured with the same ROI, and the signal to noise ratio (SNR) and
contrast to noise ratio (CNR) were calculated. And subjective evaluation of the image were subjectively evaluated. Results There
was no significant difference in CT value of reconstructed images among the 6 groups (P>0.05). The SD, SNR and CNR of AIIR
grade 1 to 5 images were better than those of Karl group, and SD increased with the increase of reconstruction grade, while SNR
and CNR decreased correspondingly (P<0.05). AIIR 5 had the highest subjective score of overall image quality, and “wax figure”
was acceptable, and there was a significant difference between the subjective score of AIIR 5 and Karl 5 algorithm (P<0.05).
Conclusion AIIR reconstruction algorithm can effectively reduce the image noise of aortic CTA and improve the signal-to-noise
ratio and contrast-to-noise ratio of vessels under the condition of low-dose scanning. However, among the five AIIR algorithms, the
subjective score of AIIR grade 5 image quality is higher, and it is recommended to use AIIR grade 5 reconstruction image in lowdose
aortic CTA scanning.