Comparison of image quality of 40 keV virtual monoenergetic images of vertebral arteries using DLIR and ASIR-V algorithms under a dual-low scanning protocol

IF 3.2 3区 医学 Q1 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING
Ying Tang , Haini Zhang , Lihui Chen , Meng Yu , He Zhang , Dapeng Zhang , Yang Wu , Zhongxiao Liu , Aiyun Sun , Yankai Meng , Kai Xu
{"title":"Comparison of image quality of 40 keV virtual monoenergetic images of vertebral arteries using DLIR and ASIR-V algorithms under a dual-low scanning protocol","authors":"Ying Tang ,&nbsp;Haini Zhang ,&nbsp;Lihui Chen ,&nbsp;Meng Yu ,&nbsp;He Zhang ,&nbsp;Dapeng Zhang ,&nbsp;Yang Wu ,&nbsp;Zhongxiao Liu ,&nbsp;Aiyun Sun ,&nbsp;Yankai Meng ,&nbsp;Kai Xu","doi":"10.1016/j.ejrad.2025.112276","DOIUrl":null,"url":null,"abstract":"<div><h3>Objective</h3><div>This study aims to compare the image quality of 40 keV Virtual Monoenergetic Images (VMI) of the vertebral artery (VA) reconstructed using two algorithms, Deep Learning Image Reconstruction (DLIR) and Adaptive Statistical Iterative Reconstruction (ASIR-V), under a dual-low scanning protocol, which simultaneously reduces both the radiation dose and the contrast agent dose during the CT scanning process.</div></div><div><h3>Methods</h3><div>A total of 88 patients were randomly assigned to an experimental group (n = 44) and a control group (n = 44). Both groups underwent dual-energy carotid CT angiography (DE-CTA) with a contrast agent dose of 0.5 ml/kg. The experimental group [Noise Index (NI) = 11.0] used DLIR at high-strength (DLIR-H) and medium-strength setting (DLIR-M), along with Adaptive Statistical Iterative Reconstruction algorithm set at 50 % strength (ASIR-V 50 %), while the control group (NI = 4.0) used ASIR-V 50 %. Image quality was evaluated both objectively [using CT values, noise, Signal-to-Noise Ratio (SNR), and Contrast-to-Noise Ratio (CNR)] and subjectively [using a 5-point scale for image noise, vascular contours, curve planar reformation (CPR), and overall image quality].</div></div><div><h3>Results</h3><div>The experimental group demonstrated a 44.4 % reduction in effective radiation dose compared to the control group (0.85 mSv and 1.53 mSv, respectively, <em>P</em> &lt; 0.001). DLIR algorithms, especially DLIR-H, significantly reduced image noise and improved both SNR and CNR compared to ASIR-V 50 % (<em>P</em> &lt; 0.01). Subjective evaluation revealed that more than 70 % of the images in the experimental group scored above 4 points, indicating excellent image quality. Furthermore, DLIR-H outperformed both ASIR-V 50 % and DLIR-M in both objective and subjective image quality assessments (<em>P</em> &lt; 0.01).</div></div><div><h3>Conclusion</h3><div>The DLIR algorithm, particularly DLIR-H, significantly improves the quality of 40 keV VMI of the vertebral artery under a dual-low scanning protocol. DLIR-H provides superior image quality with reduced radiation exposure, making it a promising option for clinical applications in patients with posterior circulation ischemic stroke. These findings are important for clinical practice as they suggest that DLIR-H can enhance diagnostic accuracy while minimizing risks associated with radiation exposure, particularly in vulnerable patient populations.</div></div>","PeriodicalId":12063,"journal":{"name":"European Journal of Radiology","volume":"191 ","pages":"Article 112276"},"PeriodicalIF":3.2000,"publicationDate":"2025-07-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"European Journal of Radiology","FirstCategoryId":"3","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0720048X25003626","RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING","Score":null,"Total":0}
引用次数: 0

Abstract

Objective

This study aims to compare the image quality of 40 keV Virtual Monoenergetic Images (VMI) of the vertebral artery (VA) reconstructed using two algorithms, Deep Learning Image Reconstruction (DLIR) and Adaptive Statistical Iterative Reconstruction (ASIR-V), under a dual-low scanning protocol, which simultaneously reduces both the radiation dose and the contrast agent dose during the CT scanning process.

Methods

A total of 88 patients were randomly assigned to an experimental group (n = 44) and a control group (n = 44). Both groups underwent dual-energy carotid CT angiography (DE-CTA) with a contrast agent dose of 0.5 ml/kg. The experimental group [Noise Index (NI) = 11.0] used DLIR at high-strength (DLIR-H) and medium-strength setting (DLIR-M), along with Adaptive Statistical Iterative Reconstruction algorithm set at 50 % strength (ASIR-V 50 %), while the control group (NI = 4.0) used ASIR-V 50 %. Image quality was evaluated both objectively [using CT values, noise, Signal-to-Noise Ratio (SNR), and Contrast-to-Noise Ratio (CNR)] and subjectively [using a 5-point scale for image noise, vascular contours, curve planar reformation (CPR), and overall image quality].

Results

The experimental group demonstrated a 44.4 % reduction in effective radiation dose compared to the control group (0.85 mSv and 1.53 mSv, respectively, P < 0.001). DLIR algorithms, especially DLIR-H, significantly reduced image noise and improved both SNR and CNR compared to ASIR-V 50 % (P < 0.01). Subjective evaluation revealed that more than 70 % of the images in the experimental group scored above 4 points, indicating excellent image quality. Furthermore, DLIR-H outperformed both ASIR-V 50 % and DLIR-M in both objective and subjective image quality assessments (P < 0.01).

Conclusion

The DLIR algorithm, particularly DLIR-H, significantly improves the quality of 40 keV VMI of the vertebral artery under a dual-low scanning protocol. DLIR-H provides superior image quality with reduced radiation exposure, making it a promising option for clinical applications in patients with posterior circulation ischemic stroke. These findings are important for clinical practice as they suggest that DLIR-H can enhance diagnostic accuracy while minimizing risks associated with radiation exposure, particularly in vulnerable patient populations.
双低扫描方案下DLIR和ASIR-V算法40 keV椎动脉虚拟单能量图像的图像质量比较
目的比较深度学习图像重建(DLIR)和自适应统计迭代重建(ASIR-V)两种算法在双低扫描方案下重建的40 keV椎动脉(VA)虚拟单能图像(VMI)的图像质量,同时降低CT扫描过程中的辐射剂量和造影剂剂量。方法将88例患者随机分为实验组(n = 44)和对照组(n = 44)。两组均行双能颈动脉CT血管造影(DE-CTA),造影剂剂量为0.5 ml/kg。实验组[噪声指数(NI) = 11.0]使用DLIR在高强度(DLIR- h)和中强度(DLIR- m)设置,并使用自适应统计迭代重建算法设置在50%强度(ASIR-V 50%),对照组(NI = 4.0)使用ASIR-V 50%。对图像质量进行客观评价[使用CT值、噪声、信噪比(SNR)和对比噪声比(CNR)]和主观评价[使用图像噪声、血管轮廓、曲线平面重构(CPR)和整体图像质量的5分制]。结果实验组的有效辐射剂量比对照组降低44.4%(分别为0.85 mSv和1.53 mSv);0.001)。与ASIR-V相比,DLIR算法,特别是DLIR- h,显著降低了图像噪声,并将信噪比和信噪比提高了50% (P <;0.01)。主观评价显示,实验组70%以上的图像得分在4分以上,表明图像质量很好。此外,DLIR-H在客观和主观图像质量评估方面均优于ASIR-V 50%和DLIR-M (P <;0.01)。结论DLIR算法,特别是DLIR- h,在双低扫描方案下显著提高了40 keV椎动脉VMI的质量。DLIR-H提供卓越的图像质量,减少辐射暴露,使其成为后循环缺血性卒中患者临床应用的一个有希望的选择。这些发现对临床实践很重要,因为它们表明DLIR-H可以提高诊断准确性,同时最大限度地减少与辐射暴露相关的风险,特别是在脆弱的患者群体中。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
CiteScore
6.70
自引率
3.00%
发文量
398
审稿时长
42 days
期刊介绍: European Journal of Radiology is an international journal which aims to communicate to its readers, state-of-the-art information on imaging developments in the form of high quality original research articles and timely reviews on current developments in the field. Its audience includes clinicians at all levels of training including radiology trainees, newly qualified imaging specialists and the experienced radiologist. Its aim is to inform efficient, appropriate and evidence-based imaging practice to the benefit of patients worldwide.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信