肺结节术前薄层CT征象及定量参数与术后病理对照研究  

A Controlled Study of CT Signs and Quantitative Parameters and Postoperative Pathology

作　　者：高大圣 胡茂能 岳庆峰 王晶 刘云峰 GAO Da-sheng;HU Mao-neng;YUE Qing-feng;WANG Jing;LIU Yun-feng(Imaging Center of Hefei Third People's Hospital,Hefei 230022,Anhui Province,China)

机构地区：[1]安徽省合肥市第三人民医院影像中心,安徽合肥230022

出　　处：《中国CT和MRI杂志》2023年第6期58-60,共3页Chinese Journal of CT and MRI

基　　金：多模态影像学精准诊断尘肺病的研发(J2019Y06)。

摘　　要：目的通过肺结节术前薄层CT及术后病理结果对照分析,总结薄层CT征象及定量参数与病变性质的相关性。方法选取2019年1月至2022年1月我院孤立性肺结节(SPN)患者100例,术前均行薄层CT扫描,术后取得病理结果。比较恶性SPN患者与良性SPN患者一般临床特点、薄层CT征象及定量参数,并分析人工智能(AI)辅助薄层CT对SPN的定性诊断价值。结果100例SPN患者均经手术病理证实,恶性59例为恶性组,良性41例为良性组;恶性组吸烟史比例(45.76%)高于良性组(24.39%),年龄大于良性SPN患者(P<0.05);征象中,恶性组分叶征、毛刺征、血管集束征、位于肺上叶比例在恶性组中的发生率较良性组高,空洞征比例低于良性组(P<0.05),量化参数中,恶性组分叶征象>3比例、毛刺数>3比例、结节直径8 mm﹤D≤30 mm比例高于良性组(P<0.05);AI辅助薄层CT定性诊断SPN的敏感度为79.66%(47/59),特异度为90.24%(37/41),准确度为84.00%(84/100)。结论薄层CT征象及征象定量、半定量分析是SPN良恶性鉴别诊断的关键,AI辅助薄层CT在SPN定性诊断中具有重要应用价值。Objective To summarize the correlation between the thin-slice CT signs and quantitative parameters and the nature of the lesion through the comparative analysis of preoperative thin-slice CT and postoperative pathological results of pulmonary nodules,in order to provide a basis for better guidance of preoperative evaluation.Methods A total of 100 patients with solitary pulmonary nodules(SPN)in our hospital from January 2019 to January 2022 were selected.Thin-slice CT scans were performed before operation,and pathological results were obtained after operation.The general clinical features,thin-slice CT signs and quantitative parameters of malignant SPN patients and benign SPN patients were compared,and the qualitative diagnosis value of artificial intelligence(AI)-assisted thin-slice CT for SPN was analyzed.Results 100 SPN patients were confirmed by surgery and pathology,59 malignant patients were classified as malignant group and 41 benign patients as benign group;the proportion of smoking history in malignant group(45.76%)was higher than that in benign group(24.39%),and the age was older than that of benign SPN patients(P<0.05);among the signs,the malignant component of lobe sign,spiculation sign,vascular bundle sign,and the proportion located in the upper lobe of the lung were higher than those of the benign group,and the proportion of cavity sign was lower than that of the benign group(P<0.05).Among the quantitative parameters,the malignant component The proportion of leaf signs>3,the number of burrs>3,and the nodule diameter of 8 mm<D≤30 mm were higher than those of the benign group(P<0.05);the sensitivity of AI-assisted thin-slice CT for qualitative diagnosis of SPN was 79.66%(47/59),the specificity was 90.24%(37/41),and the accuracy was 84.00%(84/100).Conclusion Thin-slice CT signs and quantitative and semi-quantitative analysis of signs are the key to differential diagnosis of benign and malignant SPN,and AI-assisted thin-slice CT has important application value in the qualitative diagnosis of SPN.

关 键 词：肺结节 薄层CT 征象 定量参数 术后病理 

分 类 号：R445.2[医药卫生—影像医学与核医学]