
浏览全部资源
扫码关注微信
网络出版:2021-10-28,
纸质出版:2021-10-28
移动端阅览
李宁,阚艳敏,李晓松,等. 基于SWE及CEUS的甲状腺乳头状癌诊断模型的建立[J]. 肿瘤影像学, 2021, 30(5): 392-398 https://doi.
org/10.19732/j.cnki.2096-6210.2021.05.012
李宁,阚艳敏,李晓松,等. 基于SWE及CEUS的甲状腺乳头状癌诊断模型的建立[J]. 肿瘤影像学, 2021, 30(5): 392-398 https://doi. DOI: 10.19732/j.cnki.2096-6210.2021.05.012.
org/10.19732/j.cnki.2096-6210.2021.05.012 DOI:
目的:
建立基于实时剪切波弹性成像(shear wave elastography,SWE)及超声造影(contrast-enhanced ultrasound,CEUS)的甲状腺乳头状癌(papillary carcinoma of the thyroid,PTC)诊断模型,为临床诊断提供重要依据。
方法:
选取2018年9月2020年3月于华北理工大学附属医院住院拟行甲状腺结节切除术的患者79例,共85个结节,随机分为建模组(72个结节)与验证组(13个结节)。建模组结节依据术后病理学检查结果分为PTC组(35个)和非PTC组(37个)。所有患者均行SWE及CEUS检查,并记录结节的弹性最大值(maximum elasticity,Emax)、最小值(minimum elasticity,Emin)、均值(mean value,Emean)、标准差(standard deviation,Esd)、与周围正常组织的比值(ratio to normal surrounding tissue,Eratio)、造影增强程度、增强特点、造影剂分布、与周边甲状腺实质进入及消退时间比较、结节的达峰浓度(peak concentration,Peak)、达峰时间(peak time,TP)、时间-强度曲线下面积(area of time-intensity curve,AUCt)、平均渡越时间(mean transit time,MTT)。比较PTC组与非PTC组超声指标的差异;采用Fisher 逐步判别分析法建立基于超声特征的PTC诊断模型,并进行验证,应用受试者工作特征(receiver operating characteristic ROC)曲线对其诊断效能进行评价。
结果:
两组之间,结节Emax、Emin、Emean、Esd、Eratio、增强程度、增强特点、造影剂分布、与周边甲状腺组织进入时间的比较、Peak、AUCt差异均有统计学意义(
P
<0.05);采用Fisher逐步判别分析法建立的诊断模型为PTC=-4.182+1.126
X
1
+2.556
X
3
+0.001X 7 ,非PTC=-6.596+0.416
X
1
+4.576
X
3
+0.002
X
7
(
X
1
、
X
3
、
X
7
分别代表Eratio、增强程度、AUCt),对模型进行自身验证、交叉验证及验证组结节验证,正确率分别为81.9%、79.2%和84.6%,ROC曲线分析显示其灵敏度及特异度分别为82.9%、79.1%。
结论:
采用Fisher 逐步判别分析法建立的基于SWE及CEUS的PTC诊断模型可以为临床早期诊断及干预提供客观影像学依据,有较高的应用价值。
Objective:
To establish a diagnostic model of papillary carcinoma of the thyroid (PTC) based on the shear wave elastography (SWE) and contrast-enhanced ultrasound (CEUS)
and provide an important basis for diagnosis of clinical.
Methods:
A total of 79 patients with 85 nodules who were hospitalized in Affiliated Hospital of North China University of Technology from September 2018 to March 2020 were selected and randomly divided into 72 nodules in the modeling group and 13 nodules in the verification group. The modeling group were divided into PTC group and non-PTC group according to postoperative pathological results. All patients underwent SWE and CEUS
and maximum elasticity (Emax)
minimum elasticity (Emin)
mean value (Emean)
standard deviation (Esd)
ratio to normal surrounding tissue (Eratio)
contrast enhancement degree
enhancement characteristics
distribution of contrast agent
comparison with peripheral thyroid parenchyma enhancement and regression time
peak concentration of nodules (Peak)
peak time (TP)
area of time-intensity curve (AUCt)
mean transit time (MTT) were recorded. The differences ofultrasonic indexes between PTC group and non PTC group were compared; the diagnostic model of PTC based on ultrasonic features was established by Fisher stepwise discriminant analysis and validated
receiver operating characteristic (ROC) curve was used to evaluate the diagnostic efficiency.
Results:
There were statistically significant differences in Emax
Emin
Emean
Esd
Eratio
enhancement degree
enhancement characteristics
distribution of contrast medium
enhancement time with peripheral thyroid tissue
Peak and AUCt between the two groups (
P
<0.05); the diagnostic mod
el was established [PTC=-4.182+1.126
X
1
+2.556
X
3
+0.001
X
7
non-PTC=-6.596+0.416
X
1
+4.576
X
3
+0.002
X
7
(
X
1
X
3
X
7
represent Eratio
enhancement degree
AUCt)]
the accuracies of self verification
cross validation and external data validation were 81.9%
79.2% and 84.6%
ROC curve analysis showed that the sensitivity and specificity were 82.9% and 79.1%.
Conclusion:
The diagnostic model of PTC based on SWE and CEUS established by Fisher stepwise discriminant analysis can provide objective imaging basis for early diagnosis and intervention of clinical
and has certain application value.
0
浏览量
1471
下载量
1
CNKI被引量
关联资源
相关文章
相关作者
相关机构
京公网安备11010802024621