Lucero-Garófano Álvaro, Aliena-Valero Alicia, Vielba-Gómez Isabel, Escudero-Martínez Irene, Morales-Caba Lluís, Aparici-Robles Fernando, Tarruella Hernández Diana L, Fortea Gerardo, Tembl José I, Salom Juan B, Manjón José V
Unidad Mixta de Investigación Cerebrovascular, Instituto de Investigación Sanitaria La Fe, Valencia, Spain.
Instituto de Aplicaciones de las Tecnologías de la Información y de las Comunicaciones Avanzadas (ITACA), Universitat Politècnica de València, Valencia, Spain.
Front Neurol. 2025 Jan 17;16:1534845. doi: 10.3389/fneur.2025.1534845. eCollection 2025.
Etiological classification of ischemic stroke is fundamental for secondary prevention, but frequently results in undetermined cause. We aimed to develop a Deep Learning (DL)-based model for automatic etiological classification of ischemic stroke using digital images of thrombi retrieved by mechanical thrombectomy.
Patients with large vessel occlusion stroke subjected to mechanical thrombectomy between April 2016 and January 2023 at La Fe University and Polytechnic Hospital in Valencia were included. Thrombus digital images were obtained and clinical characteristics, including TOAST etiological classification as reference standard, were retrieved. Statistical analysis was performed to compare clinical characteristics between atherothrombotic and cardioembolic strokes. A DL method was designed based on two deep neural networks for: (1) image segmentation and (2) image classification including clinical characteristics. The metrics used were DICE coefficient for the segmentation network, and accuracy, precision, sensitivity, specificity and area under the curve (AUC) for the predictions of the classification network.
A total of 166 patients (mean age 69 [SD, 13], 67 female) were included. TOAST classification was: 31 atherothrombotic, 87 cardioembolic, and 48 cryptogenic. The segmentation network achieved an average DICE coefficient of 0.96 [SD, 0.13]. The optimal fused imaging and clinical classification network had a 0.968 accuracy [95% CI, 0.935-0.994], and AUC of 0.947 [95% CI, 0.870-1]. Cryptogenic thrombi were classified as cardioembolic (96%) or atherothrombotic (4%).
Two convolutional neural networks perform the automatic segmentation of thrombus images and, combined with selected clinical characteristics, their accurate and precise classification into atherothrombotic or cardioembolic etiology in patients with acute ischemic stroke.
缺血性卒中的病因分类是二级预防的基础，但常常导致病因不明。我们旨在开发一种基于深度学习（DL）的模型，通过机械取栓获取的血栓数字图像对缺血性卒中进行自动病因分类。
纳入2016年4月至2023年1月在巴伦西亚拉费大学理工医院接受机械取栓的大动脉闭塞性卒中患者。获取血栓数字图像，并检索包括TOAST病因分类作为参考标准的临床特征。进行统计分析以比较动脉粥样硬化血栓形成性卒中和心源性栓塞性卒中的临床特征。基于两个深度神经网络设计了一种深度学习方法，用于：（1）图像分割和（2）包括临床特征的图像分类。使用的指标为分割网络的DICE系数，以及分类网络预测的准确率、精确率、灵敏度、特异度和曲线下面积（AUC）。
共纳入166例患者（平均年龄69岁[标准差，13]，女性67例）。TOAST分类为：动脉粥样硬化血栓形成性31例，心源性栓塞性87例，隐源性48例。分割网络的平均DICE系数为0.96[标准差，0.13]。最佳融合成像和临床分类网络的准确率为0.968[95%可信区间，0.935 - 0.994]，AUC为0.947[95%可信区间，0.870 - 1]。隐源性血栓被分类为心源性栓塞性（96%）或动脉粥样硬化血栓形成性（4%）。
两个卷积神经网络对血栓图像进行自动分割，并结合选定的临床特征，对急性缺血性卒中患者的动脉粥样硬化血栓形成性或心源性栓塞性病因进行准确和精确分类。
