Department of Surgery, Amsterdam University Medical Centers location, Vrije Universiteit, Amsterdam, The Netherlands.
Amsterdam Cardiovascular Sciences, Amsterdam, the Netherlands.
J Endovasc Ther. 2023 Dec;30(6):822-827. doi: 10.1177/15266028221105840. Epub 2022 Jul 9.
Modern endovascular hybrid operating rooms generate large amounts of medical images during a procedure, which are currently mostly assessed by eye. In this paper, we present fully automatic segmentation of the stent graft on the completion digital subtraction angiography during endovascular aneurysm repair, utilizing a deep learning network.
Completion digital subtraction angiographies (cDSAs) of 47 patients treated for an infrarenal aortic aneurysm using EVAR were collected retrospectively. A two-dimensional convolutional neural network (CNN) with a U-Net architecture was trained for segmentation of the stent graft from the completion angiographies. The cross-validation resulted in an average Dice similarity score of 0.957 ± 0.041 and median of 0.968 (IQR: 0.950 - 0.976). The mean and median of the average surface distance are 1.266 ± 1.506 mm and 0.870 mm (IQR: 0.490 - 1.430), respectively.
We developed a fully automatic stent graft segmentation method based on the completion digital subtraction angiography during EVAR, utilizing a deep learning network. This can provide the platform for the development of intraoperative analytical applications in the endovascular hybrid operating room such as stent graft deployment accuracy, endoleak visualization, and image fusion correction.
现代血管内杂交手术室在手术过程中会生成大量的医学图像，目前这些图像主要通过肉眼进行评估。在本文中，我们提出了一种利用深度学习网络自动分割血管内动脉瘤修复术（EVAR）完成数字减影血管造影（cDSA）中覆膜支架的方法。
回顾性收集了 47 例接受腹主动脉下段 EVAR 治疗的患者的 cDSA 图像。采用二维卷积神经网络（CNN）的 U-Net 架构对支架进行分割。交叉验证得到平均 Dice 相似系数为 0.957±0.041，中位数为 0.968（IQR：0.950-0.976）。平均表面距离的均值和中位数分别为 1.266±1.506mm 和 0.870mm（IQR：0.490-1.430）。
我们开发了一种基于 EVAR 完成数字减影血管造影的全自动覆膜支架分割方法，利用深度学习网络。这可为血管内杂交手术室中的术中分析应用（如支架释放精度、内漏可视化和图像融合校正）的开发提供平台。
