HRCT平扫加CT直方图对混合磨玻璃结节的良恶性鉴别诊断
Differential diagnosis of benign and malignant mixed-ground glass nodules by HRCT plain scan and CT histogram
刘凡萍 1封俊 1李清婉 1阮进秋 1蒋赵娟 1高德培2
作者信息
- 1. 云南省昭通市第一人民医院放射科 云南 昭通 657000
- 2. 云南省肿瘤医院放射科 云南 昆明 650118
- 折叠
摘要
目的 探讨通过术前高分辨率CT(HRCT)平扫的表现加CT灰度直方图信息对混合性磨玻璃结节(mixed ground-glass nodules,mGGN)良恶性鉴别诊断的价值.方法 选取经病理证实的 117 例胸部HRCT检查表现为mGGN的患者,根据病理结果分为良性结节组(43 例)和恶性结节组(74 例).分析两组结节整体影像学信息、结节内实性成分、结节直方图信息.结果 良恶性结节的大小、体积、形态、毛刺征等整体影像学信息以及结节内实性部分的形状和占比在两组间的差异具有统计学意义(P<0.05);CT灰度直方图中参数比较在两组间差异有统计学意义(P<0.05);受试者工作特征曲线(receiver operating characteristic curve,ROC)分析熵的曲线下面积最大为 0.770,提示其预测效能最高.结论 HRCT平扫常规图像信息加上CT灰度直方图信息能够鉴别mGGN的良恶性,可以帮助临床医师对mGGN提供治疗决策.
Abstract
Objective To explore the value of preoperative high-resolution CT plain scan image information and CT gray his-togram information in the differential diagnosis of benign and malignant mixed-ground glass nodules(mGGN).Methods The CT plain images data of 117 patients with mGGN confirmed by pathology were collected.They were divided into benign nodule group with 43 cases and malignant nodule group with 74 cases.The overall image information of the two groups of nodules,the image information of solid components in the nodules and the histogram information of the nodules were analyzed.Results The overall image information of benign and malignant nodules,such as size,volume,shape,spicule sign,and the shape and pro-portion of solid components of nodules were statistically significant between the two groups(P<0.05);The parameters in CT gray histogram were significantly different between the two groups(P<0.05);The maximum area under the curve of entropy of ROC analysis was 0.770,indicating that the prediction efficiency was the highest.Conclusion The image information ofcon-ventional HRCT plain scan combined with CT gray histogram information can identify the benign and malignantmGGN,which can help clinicians make treatment decisions for mGGN.
关键词
混合磨玻璃结节/直方图/肺癌/体层摄影术,X线计算机Key words
Mixed ground glass nodules/Histogram/Lung cancer/Tomography,X-ray computed引用本文复制引用
出版年
2024
NETL
NSTL
万方数据