深度学习图像重建与自适应统计迭代重建在冠心病高危人群冠状动脉钙化评分中的应用

Deep learning image reconstruction and adaptive statistical iterative reconstruction on coronary artery calcium scoring in high risk population for coronary heart disease.

作者信息

Zhu Lijuan, Shi Xiaomeng, Tang Lusong, Machida Haruhiko, Yang Lili, Ma Meixiang, Ha Ruoshui, Shen Yun, Wang Fang, Chen Dazhi

机构信息

People's Hospital of Ningxia Hui Autonomous Region to People's Hospital of Ningxia Hui Autonomous Region, Ningxia Medical University, 301 Zhengyuan North Street, Jinfeng District, Yinchuan, Ningxia, China.

GE (China) CT Imaging Research Center, Shanghai, China.

出版信息

BMC Med Inform Decis Mak. 2025 Jul 1;25(1):212. doi: 10.1186/s12911-025-03049-w.

DOI:10.1186/s12911-025-03049-w

PMID:40598159

原文链接:https://pmc.ncbi.nlm.nih.gov/articles/PMC12210892/

Abstract

OBJECTIVE

Deep learning image reconstruction (DLIR) technology effectively improves the image quality while maintaining spatial resolution. The impact of DLIR on the quantification of coronary artery calcium (CAC) is still unclear. The purpose of this study was to investigate the effect of DLIR on the quantification of coronary calcium in high-risk populations.

METHODS

A retrospective study was conducted on patients who underwent coronary artery CT angiography (CCTA) at our hospital(China) from February 2022 to September 2022. Raw data were reconstructed with filtered back projection (FBP) reconstruction, 40% and 80% level adaptive statistical iterative reconstruction-veo (ASiR-V 40%, ASiR-V 80%) and low, medium and high-level deep learning algorithm (DLIR-L, DLIR-M, and DLIR-H). Calculate and compare the signal-to-noise and contrast-to-noise ratio, volumetric score, mass scores, and Agaston score of 6 sets of images.

RESULTS

There were 178 patients, female (107), mean age (62.43 ± 9.26), and mean BMI (25.33 ± 3.18) kg/m. Compared with FBP, the image noise of ASiR-V and DLIR was significantly reduced (P < 0.001). There was no significant difference in Agaston score, volumetric score, and mass scores among the six reconstruction algorithms (all P > 0.05). Bland-Altman diagram indicated that the Agatston scores of the five reconstruction algorithms showed good agreement with FBP, with DLIR-L(AUC, 110.08; 95% CI: 26.48, 432.92;)and ASIR-V40% (AUC,110.96; 95% CI: 26.23, 431.34;) having the highest consistency with FBP.

CONCLUSION

Compared with FBP, DLIR and ASiR-V improve CT image quality to varying degrees while having no impact on Agatston score-based risk stratification.

CLINICAL RELEVANCE STATEMENT

CACS is a powerful tool for cardiovascular risk stratification, and DLIR can improve image quality without affecting CACS, making it widely applicable in clinical practice.

摘要

目的

深度学习图像重建（DLIR）技术在保持空间分辨率的同时有效提高了图像质量。DLIR对冠状动脉钙化（CAC）定量的影响尚不清楚。本研究的目的是探讨DLIR对高危人群冠状动脉钙化定量的影响。

方法

对2022年2月至2022年9月在我院（中国）接受冠状动脉CT血管造影（CCTA）的患者进行回顾性研究。原始数据采用滤波反投影（FBP）重建、40%和80%水平的自适应统计迭代重建-veo（ASiR-V 40%，ASiR-V 80%）以及低、中、高水平深度学习算法（DLIR-L、DLIR-M和DLIR-H）进行重建。计算并比较6组图像的信噪比、对比噪声比、体积评分、质量评分和阿加斯顿评分。

结果

共178例患者，女性（107例），平均年龄（62.43±9.26）岁，平均体重指数（25.33±3.18）kg/m²。与FBP相比，ASiR-V和DLIR的图像噪声显著降低（P<0.001）。六种重建算法的阿加斯顿评分、体积评分和质量评分之间无显著差异（均P>0.05）。Bland-Altman图表明，五种重建算法的阿加斯顿评分与FBP显示出良好的一致性，其中DLIR-L（AUC，110.08；95%CI：26.48，432.92；）和ASIR-V40%（AUC，110.96；95%CI：26.23，431.34；）与FBP的一致性最高。