原发三叉神经痛发病预警模型构建及3D-FIESTACISS图像在其治疗中的应用前景  

Application Prospects of Original Troll Neural Pain Sick Early Warning Model and 3D-Fiestaciss Image in its Treatment

作　　者：娄晶[1] 魏军伟 易明岗[3] Lou jing;Wei Junwei;Yi Ming-gang(Department of Imaging,Jinan Hospital of Traditional Chinese Medicine,Shandong 250012,China;Imaging Department of Changyi People’s Hospital,Shandong 261300,China;Department of Imaging,Jinan Children’s Hospital,Shandong 250000,China)

机构地区：[1]济南市中医医院影像科,山东250012 [2]昌邑市人民医院影像科,山东261300 [3]济南市儿童医院影像科,山东250000

出　　处：《影像技术》2021年第3期4-10,共7页Image Technology

基　　金：济南市卫计委青年科技计划项目(2017-2-17);山东省医药卫生科技发展计划项目(2017WS530)。

摘　　要：目的:探讨原发三叉神经痛(PTN)发病预警模型构建及3D-FIESTACISS图像在其治疗中的应用前景。方法:回顾性分析我院收治的160例确诊为PTN患者的临床资料并建立PTN患者发病的列线图预测模型;比较常见的MRI检测方法与3D-FIESTACISS图像在判断PTN患者神经压迫程度中的诊断情况。结果:病程>3年、发病类型为典型、术前血管有压迫、疼痛区域为复合支、不充分的影像学检查是影响PTN发病的独立危险因素(P<0.05)。筛选出危险因素建立列线图预测模型验证结果显示,建模集和验证集的C-index指数分别为0.864和0.868,ROC曲线下面积分别为0.864和0.868,具有良好的预测精准度。与3D-TOF-MRA、3D-FIESTA图像结果相比,3D-FIESTACISS图像的准确率最高(P=0.963)。3D-TOF-MRA、3D-FIESTA及3D-FIESTACISS 3种方法的AUC分别为0.755、0.790、0.899,预测概率AUC为0.928(P<0.05)。结论:基于危险因素建立的列线图模型可有效预测和量化PTN发病风险,临床可在早期采取干预措施,改善PTN患者的复发率及预后。3D-FIESTACISS在术前诊断PTN中具有较高的准确性,可用于指导PTN患者术前的责任区血管的探查。Objective:To investigate the development of early warning model of primary trigeminal neuralgia(PTN)and the application prospect of 3D-Fiestaciss image in its treatment.Methods:The clinical data of 160 patients diagnosed with PTN in our hospital were retrospectively analyzed and a prediction model of the incidence of PTN patients was established.The diagnosis of common MRI detection methods and 3D-Fiestaciss images in determining the degree of nerve compression in PTN patients.Results:>3 years of course,typical type of disease,preoperative vascular compression,complex branches of pain area,and inadequate imaging examination were independent risk factors for PTN(P<0.05).The validation results of the prediction model of the histogram established by screening out the risk factors showed that the C-index of the modeling set and the verification set were 0.864 and 0.868,respectively,and the area under the ROC curve was 0.864 and 0.868,respectively,showing good prediction accuracy.Compared with 3D-TOF-MRA and 3D-FIESTA images,the accuracy of 3D-FIESTACISS images was the highest(P=0.963).The AUC of 3D-TOF-MRA,3D-FIESTA and 3D-FIESTACISS were 0.755,0.790 and 0.899,respectively,and the predictive probability AUC was 0.928(P<0.05).Conclusion:The histogram model based on risk factors can effectively predict and quantify the risk of PTN.Clinically,intervention measures can be taken for risk factors in the early stage to improve the recurrence rate and prognosis of PTN patients.3D-Fiestaciss has a high accuracy in preoperative diagnosis of PTN,and can be used to guide preoperative vascular exploration in the responsible area of PTN patients.

关 键 词：原发三叉神经痛 复发 危险因素 3D-FIESTACISS 

分 类 号：R714[医药卫生—妇产科学]