Abstract：
The vertebral artery\'s morphological characteristics are crucial in spontaneous vertebral artery dissection (sVAD). We aimed to investigate morphologic features related to ischemic stroke (IS) and develop a novel prediction model. Out of 126 patients, 93 were finally analyzed. We constructed 3D models and morphological analyses. Patients were randomly classified into training and validation cohorts (3:1 ratio). Variables selected by LASSO - including five morphological features and five clinical characteristics - were used to develop prediction model in the training cohort. The model exhibited a high area under the curve (AUC) of 0.944 (95%CI, 0.862-0.984), with internal validation confirming its consistency (AUC = 0.818, 95%CI, 0.597-0.948). Decision curve analysis (DCA) indicated clinical usefulness. Morphological features significantly contribute to risk stratification in sVAD patients. Our novel developed model, combining interdisciplinary parameters, is clinically useful for predicting IS risk. Further validation and in-depth research into the hemodynamics related to sVAD are necessary.
摘要：
椎动脉形态特征在自发性椎动脉夹层(sVAD)中至关重要。我们旨在研究与缺血性卒中（IS）相关的形态学特征，并开发一种新的预测模型。126名患者中,最后分析了93个。我们构建了三维模型和形态学分析。将患者随机分为训练和验证队列(3:1比例)。LASSO选择的变量-包括五个形态学特征和五个临床特征-用于在训练队列中开发预测模型。该模型显示出较高的曲线下面积（AUC）为0.944（95CI，0.862-0.984），内部验证确认其一致性(AUC=0.818,95CI,0.597-0.948)。决策曲线分析(DCA)提示临床有用性。形态学特征显著有助于sVAD患者的风险分层。我们新开发的模型，结合跨学科参数，在临床上对预测IS风险是有用的。需要进一步验证和深入研究与sVAD相关的血液动力学。
