期刊文献+

能谱CT虚拟平扫及水基图定量参数在诊断颅脑血管内治疗术后颅内出血的研究

The Value of Quantitative Parameters on Virtual Non-Contrast and Water-Based Images of Brain Spectral CT in Early Diagnosing Intracranial Hemorrhage after Endovascular Treatment
原文传递
导出
摘要 目的 探讨能谱CT虚拟平扫(VNC)和水(碘)基图在颅脑血管内治疗后诊断出血(ICH)的价值。方法回顾性分析69例存在大血管狭窄或闭塞且接受血管内治疗、术后采用能谱CT扫描的患者,以24 h复查脑平扫CT判断有无脑出血为金标准分为出血组和非出血组。分别测量VNC图的CT值和水(碘)图的密度值。使用SPSS分析水基图与VNC定量参数诊断术后出血转化的一致性和相关性,并进一步通过受试者工作特征(ROC)曲线分析其诊断效能。由2名医师采用盲法分别单独用VNC图、VNC结合水基图进行诊断信心评分。诊断信心评分采用配对样本t检验。结果 基于最佳截断值为1043.9 mg/cm^(3)的水(碘)基图水密度值和最佳截断值为52.9 HU的VNC图CT值的诊断结果具有较强的一致性(k=0.964,P<0.001)。经双变量Spearman秩和检验结果显示,上述水密度值和VNC CT值具有显著相关性(r_(s)=0.967,P<0.001)。ROC曲线分析表明,水基图和VNC的曲线下面积(AUC)值分别为0.974(95%CI:0.904~0.997),0.976(95%CI:0.906~0.998),差异无统计学意义(P>0.05),最佳截断值下水基图诊断ICH的敏感度、特异度、阳性预测值、阴性预测值、准确率分别为94.7%、98.0%、94.7%、98.0%、97.1%。最佳截断值下VNC诊断ICH的敏感度、特异度、阳性预测值、阴性预测值、准确率分别为94.7%、96.0%、90.0%、98.0%、95.7%。2名医师对VNC结合水基图有更高的诊断信心评分(P<0.05)。结论 能谱CT水基图和VNC图定量分析对于血管内治疗后的ICH具有较高的诊断价值,最佳截断值分别为水基图水密度值1043.9 mg/cm^(3),VNC CT值52.9 HU。VNC结合水基图具有更高的诊断灵敏度,并可以提高医师的诊断信心。 Objective To explore the value of virtual non-contrast(VNC)and water-based images in the diagnosis of intracranial hemorrhage(ICH)after endovascular treatment with brain spectral CT.Methods 69 patients underwent spectral CT scanning immediately after the endovascular treatment were retrospectively enrolled in this study.Patients were divided into hemorrhagic group and non-hemorrhagic group based on the gold standards,which were 24-hour follow-up non-contrast CT examinations.The CT values and the water concentration of VNC images and water-based images were meas-ured.We had analyzed the consistency and receiver operating characteristic curves(ROC)of the diagnostic performance of VNC and water-based images.Two radiologists conducted subjective assessment on the VNC and water-based diagnostic confidence of ICH.Results The cut-off values were 52.9 HU for VNC and 1043.9 mg/cm3 for water-based images on ICH diagnose.The diagnostic results have strong consistency(k=0.964,P<0.001).The above water density and VNC CT values were significantly correlated(r_(s)=0.967,P<0.001)according to the bivariate Spearman rank-sum test.The ROC curve analysis showed that the areas under the curves of water-based images and VNC images were both 0.974(95%CI:0.904~0.997),0.976(95%CI:0.906~0.998),respectively.The difference was not statistically significant(P>0.05).The sensitivity,specificity,positive predictive value,negative predictive value,and accuracy of VNC in diagnosing ICH at the best cutoff value were 94.7%,96.0%,90.0%,98.0%,and 95.7%,respectively.The corresponding values for water density at the best cutoff were 94.7%,98.0%,94.7%,98.0%,and 97.1%,respectively.The combination of VNC and water-based images had a higher diagnostic confidence score(P<0.05).Conclusion The quantitative analysis of water-based images and VNC images using spectral CT has a high diagnostic value for imaging ICH after endovascular treatment.The best cutoff values are a water density of 1043.9 mg/cm3 and a VNC-CT value of 52.9 HU.Combining VNC and water-based images can improve diagnostic sensitivity and increase diagnostic confidence.
作者 杨志安 闵小红 徐俏宇 王落桐 韩靖 胡杨 肖焕波 顾华 YANG Zhian;MIN Xiaohong;XU Qiaoyu(Department of Radiology,Affiliated Beijing Chaoyang Hospital of Capital Medical University,Beijing 100020,P.R.China)
出处 《临床放射学杂志》 北大核心 2024年第6期872-877,共6页 Journal of Clinical Radiology
关键词 能谱CT 颅内出血 水-碘基图 虚拟平扫 血管内治疗 Spectral CT Intracranial hemorrhage Material decomposition Virtual non-contrast Endovascular Treatment postoperative
  • 相关文献

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部