From the Department of Neurological Surgery (N.I., C.-J.C.), University of Virginia Health System, Charlottesville, VA.
Department of Radiology (S. Mutasa, S. Marfatiah, A. Lignelli), Columbia University Irving Medical Center, New York.
Stroke. 2020 Mar;51(3):815-823. doi: 10.1161/STROKEAHA.119.026764. Epub 2020 Feb 12.
Background and Purpose- Perihematomal edema (PHE) is a promising surrogate marker of secondary brain injury in patients with spontaneous intracerebral hemorrhage, but it can be challenging to accurately and rapidly quantify. The aims of this study are to derive and internally validate a fully automated segmentation algorithm for volumetric analysis of PHE. Methods- Inpatient computed tomography scans of 400 consecutive adults with spontaneous, supratentorial intracerebral hemorrhage enrolled in the Intracerebral Hemorrhage Outcomes Project (2009-2018) were separated into training (n=360) and test (n=40) datasets. A fully automated segmentation algorithm was derived from manual segmentations in the training dataset using convolutional neural networks, and its performance was compared with that of manual and semiautomated segmentation methods in the test dataset. Results- The mean volumetric dice similarity coefficients for the fully automated segmentation algorithm were 0.838±0.294 and 0.843±0.293 with manual and semiautomated segmentation methods as reference standards, respectively. PHE volumes derived from the fully automated versus manual (r=0.959; <0.0001), fully automated versus semiautomated (r=0.960; <0.0001), and semiautomated versus manual (r=0.961; <0.0001) segmentation methods had strong between-group correlations. The fully automated segmentation algorithm (mean 18.0±1.8 seconds/scan) quantified PHE volumes at a significantly faster rate than both of the manual (mean 316.4±168.8 seconds/scan; <0.0001) and semiautomated (mean 480.5±295.3 seconds/scan; <0.0001) segmentation methods. Conclusions- The fully automated segmentation algorithm accurately quantified PHE volumes from computed tomography scans of supratentorial intracerebral hemorrhage patients with high fidelity and greater efficiency compared with manual and semiautomated segmentation methods. External validation of fully automated segmentation for assessment of PHE is warranted.
背景与目的- 血肿周围水肿（PHE）是自发性脑出血患者继发性脑损伤的一个很有前途的替代标志物，但准确快速地对其进行定量是具有挑战性的。本研究旨在开发并内部验证一种用于 PHE 容积分析的全自动分割算法。
方法- 将 2009 年至 2018 年间纳入颅内出血结局项目（Intracerebral Hemorrhage Outcomes Project）的 400 例连续成年自发性幕上脑出血住院患者的计算机断层扫描（CT）分为训练集（n=360）和测试集（n=40）。在训练集中，使用卷积神经网络从手动分割中推导出全自动分割算法，并在测试集中将其性能与手动和半自动分割方法进行比较。
结果- 全自动分割算法的平均体积 Dice 相似系数分别为 0.838±0.294 和 0.843±0.293，以手动和半自动分割方法为参考标准。全自动与手动（r=0.959；<0.0001）、全自动与半自动（r=0.960；<0.0001）以及半自动与手动（r=0.961；<0.0001）分割方法之间的 PHE 体积具有较强的组间相关性。全自动分割算法（平均 18.0±1.8 秒/次扫描）比手动（平均 316.4±168.8 秒/次扫描；<0.0001）和半自动（平均 480.5±295.3 秒/次扫描；<0.0001）分割方法更快地量化 PHE 体积。
结论- 与手动和半自动分割方法相比，全自动分割算法能够准确地从幕上脑出血患者的 CT 扫描中量化 PHE 体积，具有更高的保真度和效率。需要对全自动分割进行外部验证，以评估 PHE。
