BACKGROUND:Acute ischemic stroke (AIS) results in high morbidity, disability, and mortality. Early and automatic diagnosis of AIS can help clinicians administer the appropriate interventions. OBJECTIVE:To develop a deep symmetric 3D convolutional neural network (DeepSym-3D-CNN) for automated AIS diagnosis via diffusion-weighted imaging (DWI) images. METHODS:This study includes 190 study subjects (97 AIS and 93 Non-AIS) by collecting both DWI and Apparent Diffusion Coefficient (ADC) images. 3D DWI brain images are split into left and right hemispheres and input into two paths. A map with 125×253×14×12 features is extracted by each path of Inception Modules. After the features computed from two paths are subtracted through L-2 normalization, four multi-scale convolution layers produce the final predation. Three comparative models using DWI images including MedicalNet with transfer learning, Simple DeepSym-3D-CNN (each 3D Inception Module is replaced by a simple 3D-CNN layer), and L-1 DeepSym-3D-CNN (L-2 normalization is replaced by L-1 normalization) are constructed. Moreover, using ADC images and the combination of DWI and ADC images as inputs, the performance of DeepSym-3D-CNN is also investigated. Performance levels of all three models are evaluated by 5-fold cross-validation and the values of area under ROC curve (AUC) are compared by DeLong’s test. RESULTS:DeepSym-3D-CNN achieves an accuracy of 0.850 and an AUC of 0.864. DeLong’s test of AUC values demonstrates that DeepSym-3D-CNN significantly outperforms other comparative models (p < 0.05). The highlighted regions in the feature maps of DeepSym-3D-CNN spatially match with AIS lesions. Meanwhile, DeepSym-3D-CNN using DWI images presents the significant higher AUC than that either using ADC images or using DWI-ADC images based on DeLong’s test (p < 0.05). CONCLUSIONS:DeepSym-3D-CNN is a potential method for automatically identifying AIS via DWI images and can be extended to other diseases with asymmetric lesions.

中文翻译：

---

通过扩散加权图像识别急性缺血性卒中的深度对称三维卷积神经网络

背景：急性缺血性卒中 (AIS) 导致高发病率、残疾和死亡率。AIS 的早期和自动诊断可以帮助临床医生进行适当的干预。目的：开发一种深度对称 3D 卷积神经网络 (DeepSym-3D-CNN)，用于通过扩散加权成像 (DWI) 图像进行 AIS 自动诊断。方法：本研究通过收集 DWI 和表观扩散系数 (ADC) 图像，包括 190 名研究对象（97 名 AIS 和 93 名非 AIS）。3D DWI 大脑图像被分成左右半球并输入两条路径。通过 Inception Modules 的每条路径提取一个具有 125×253×14×12 特征的地图。在通过 L-2 归一化减去从两条路径计算的特征后，四个多尺度卷积层产生最终的捕食。三个使用 DWI 图像的比较模型，包括带有迁移学习的 MedicalNet、Simple DeepSym-3D-CNN（每个 3D Inception Module 被一个简单的 3D-CNN 层替换）和 L-1 DeepSym-3D-CNN（L-2 归一化被替换通过 L-1 归一化）构建。此外，使用 ADC 图像以及 DWI 和 ADC 图像的组合作为输入，还研究了 DeepSym-3D-CNN 的性能。所有三个模型的性能水平均通过 5 折交叉验证进行评估，ROC 曲线下面积 (AUC) 的值通过 DeLong 检验进行比较。结果：DeepSym-3D-CNN 的准确度为 0.850，AUC 为 0.864。DeLong 对 AUC 值的测试表明，DeepSym-3D-CNN 显着优于其他比较模型（p < 0.05）。DeepSym-3D-CNN 的特征图中突出显示的区域在空间上与 AIS 病变匹配。同时，使用 DWI 图像的 DeepSym-3D-CNN 的 AUC 显着高于使用 ADC 图像或基于 DeLong 检验的 DWI-ADC 图像（p < 0.05）。结论：DeepSym-3D-CNN 是一种通过 DWI 图像自动识别 AIS 的潜在方法，并且可以扩展到其他具有不对称病变的疾病。