| 杨贤增.高分辨率 CT的实性成分体积在肺亚实性结节浸润等级中的预测价值[J].安徽医药,2020,24(3):565-568. |
| 高分辨率 CT的实性成分体积在肺亚实性结节浸润等级中的预测价值 |
| Predictive value of volume of solid component in quantitative high resolution CT of pulmonary subsolid nodules for the pathologic grade |
| |
| DOI：10.3969/j.issn.1009?6469.2020.03.036 |
| 中文关键词: 肺肿瘤/诊断 体层摄影术，螺旋计算机 定量参数 病理学 |
| 英文关键词: Lung neoplasms/diagnosis Tomography,spiral computed Quantitative parameters Pathology |
| 基金项目: |
|
| 摘要点击次数: 1819 |
| 全文下载次数: 456 |
| 中文摘要: |
| 目的探讨高分辨率 CT的实性成分体积在预测肺亚实性结节的病理浸润等级中的价值。方法回顾性分析 2014年 5月至 2017年 10月于南阳张仲景医院经病理证实为腺癌的 86例肺亚实性结节病人的术前胸部高分辨率薄层 CT(HRCT)图像。根据病理浸润情况分为高浸润组( n＝41)和低浸润组( n＝45)，使用计算机辅助软件测量所有肺亚实性结节的肺窗实性一维长径、肺窗实性二维长径、纵隔窗实性一维长径、纵隔窗实性二维长径以及实性成分体积，采用二元 logistic回归分析得到病理浸润等级的独立预测因素并通过受试者工作特征曲线( ROC)评价其诊断效能。结果高浸润组的肺窗实性一维长径( 12.53± 4.63)mm、肺窗实性二维长径( 11.29±4.81)mm、纵隔窗实性一维长径( 8.63±2.67)mm、纵隔窗实性二维长径( 7.86±2.11)mm以及实性成分体积( 236.44±72.53)mm3显著高于低浸润组( 5.24±2.14)mm、(4.53±2.08)mm、(3.42±1.14)mm、(2.89±1.07)mm以及(73.59±18.34)mm3差异有统计学(P＜0.05)。多因素 logistic回归分析结果表明实性成分体积是肺亚实性结节病理浸润等级的独立预测OR＝1.019，95%CI:1.012~1.025)，ROC曲线分析结果表明其最佳诊断界值为 108.12 mm3诊断的灵敏度为 82.16%，特异度为 91.56%，曲线下面积为 0.912。结论高分辨率 CT定量参数对肺亚实性结节的病理浸润预测具有一意义，因素(，等级，定的价值，尤其是阈值为-300 HU的实性成分体积这一参数是其病理浸润等级的独立预测因素。 |
| 英文摘要: |
| Objective To explore the predictive value of volume of solid component in quantitative high resolution CT of pulmo?nary subsolid nodules for the pathologic grade.Methods The preoperative chest HRCT and pathological data of 86 patients withpulmonary subsolid nodules were collected from May 2014 to October 2017 in Nanyang Zhang Zhongjing Hospital，retrospectively.All specimens were divided into two groups:high infiltration group(n＝41)and low infiltration group(n＝45)according to the in?filtration grade.Computer aided diagnosis software were used to measure the one dimension maximum diameter of solid componentwith mediastinal window setting，two dimension maximum diameter of solid component with mediastinal window setting，one dimen? sion maximum diameter of whole nodule with lung window setting，two dimension maximum diameter of whole nodule with lung win? dow setting，and volume of solid component with threshold of -300 HU of all pulmonary subsolid nodules.All the quantitative fea?tures were evaluated by using univariate logistic regression analysis and multivariate logistic regression.The independent predictorsof ulmonary subsolid nodules for the pathological grade were obtained.Receiver operating characteristic(ROC)analysis was con? ducted for the independent predictive factors.Results The solid one?dimensional long diameter of the lung window was(12.53± 4.63)mm，the solid two?dimensional long diameter of the lung window was(11.29±4.81)mm，and the solid one?dimensional long diameter of the mediastinum window wa(8.63±2.67)mm，the solid two?dimensional long diameter of the mediastinum window was(7.86±2.11)mm and the area of the solid components was(236.44±72.53)mm3，they were significantly higher than those of the low infiltration group，which were(5.24±2.14)mm，(4.53±2.08)mm，(3.42±1.14)mm，(2.89±1.07)mm and 73.59±18.34)mm3，re? spectively，and the differences were statistically significant(P＜0.05).The results of multivariate logistic regression analysis showedthat the volume of solid components was an independent predictor of the pathological infiltration level of pulmonary subsolid nod?ules(OR＝1.019，95%CI:1.012?1.025)，the ROC curve analysis results show that the best diagnostic cutoff was 108.12 mm3，the diagnostic sensitivity was 82.16%，the specificity was 91.56%，and the area under the curve was 0.912.Conlcusion High resolu? tion CT quantitative parameters were valuable for predicting the pathological grade of pulmonary subsolid nodules，especially for the volume of solid component with threshold of -300 HU. |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |
