摘要
目的 探究多模态迭代重建算法(ASIR-V)、深度学习图像重建算法(DLIR)对早期肺癌患者CT成像质量的影响。方法 选取我院2020年7月~2023年3月收治的78例早期肺癌患者,分别采用50%ASIR-V、DLIR算法的中强度(DLIR-M)方法进行CT图像重建。测量并比较两种重建算法图像上胸廓入口、气管隆突下(肺部)、上腹背部的CT值和图像背景噪声(SD);由5位放射科医师对两种重建算法下的图像质量、噪声、癌结节内外部征象及癌结节分界情况进行评分。结果 50%ASIR-V、DLIR-M算法在胸廓入口、气管隆突下及上腹背部的CT值与原始图像比较差异无统计学意义(P>0.05)。DLIR-M算法在不同部位肌肉、脂肪层面的SD值显著低于原始图像和50%ASIR-V算法,差异均有统计学意义(P<0.05)。与原始图像和50%ASIR-V算法相比,DLIR-M算法图像质量、图像噪声、癌结节内外部征象及癌结节成分分界评分较高,差异均有统计学意义(P<0.05)。结论 相同扫描条件下,DLIR-M和50%ASIR-V算法的CT成像质量均优于原始图像,但DLIR-M的成像质量更佳。
Objective To investigate the effects of adaaptive statistical iterative reconstruction V(ASIR-V)and deep learning image reconstruction(DLIR)on the quality of CT images of patients with early-stage lung cancer.Methods 78 patients with early-stage lung cancer admitted to our hospital from July 2020 to March 2023 were selected for CT image reconstruction using 50%ASIR-V and DLIR algorithms with medium intensity(DLIR-M)methods,respectively.The CT values of the thoracic inlet,sub-tracheal eminence(lung),upper abdominal and dorsal regions,and the image background noise(SD)were measured and compared between the two reconstruction algorithms;five radiologists scored the image quality,noise,internal and external signs of the cancerous nodules,and the demarcation of the cancerous nodules under the two reconstruction algorithms.Results The CT values of 50%ASIR-V and DLIR-M algorithms in thoracic inlet,sub-tracheal eminence and epigastric dorsum were not statistically different from those of the original images(P>0.05),and the SD values of DLIR-M algorithm in different parts of the muscle and fat levels were significantly lower than those of the original images and 50%ASIR-V algorithm,and the differences were all statistically significant(P<0.05).Compared with the original images and the 50%ASIR-V algorithm,the DLIR-M algorithm was more effective in terms of image quality,image noise,and the internal and external signs of cancerous nodules.The image quality,image noise,internal and external signs of cancerous nodules,and the demarcation of cancerous nodule components were higher in the DLIR-M algorithm compared with the original image and the 50%ASIR-V algorithm,and the differences were statistically significant(P<0.05).Conclusion Under the same scanning conditions,the CT imaging quality of both the DLIR-M and 50%ASIR-V algorithms is superior to that of the original image,but the imaging quality of the DLIR-M is better.
作者
史同峰
杨方波
张强
孙振亚
Shi Tongfeng;Yang Fangbo;Zhang Qiang(Puyang Fifth People's Hospital,Puyang 457000,China)
出处
《华北理工大学学报（医学版）》
2023年第6期432-436,共5页
Journal of North China University of Science and Technology：Health Sciences Edition
关键词
早期肺癌
CT
算法重建
成像质量
Early-stage lung cancer
CT
Algorithmic reconstruction
Image quality
