Department of Radiology, Huashan Hospital, Fudan University, Shanghai, China.
Department of Electronic Engineering, Fudan University, Shanghai, China and Key Laboratory of Medical Imaging Computing and Computer Assisted Intervention of Shanghai, China.
J Stroke Cerebrovasc Dis. 2021 Jun;30(6):105752. doi: 10.1016/j.jstrokecerebrovasdis.2021.105752. Epub 2021 Mar 27.
To explore a new approach mainly based on deep learning residual network (ResNet) to detect infarct cores on non-contrast CT images and improve the accuracy of acute ischemic stroke diagnosis.
We continuously enrolled magnetic resonance diffusion weighted image (MR-DWI) confirmed first-episode ischemic stroke patients (onset time: less than 9 h) as well as some normal individuals in this study. They all underwent CT plain scan and MR-DWI scan with same scanning range, layer thickness (4 mm) and interlayer spacing (4 mm) (The time interval between two examinations: less than 4 h). Setting MR-DWI as gold standard of infarct core and using deep learning ResNet combined with a maximum a posteriori probability (MAP) model and a post-processing method to detect the infarct core on non-contrast CT images. After that, we use decision curve analysis (DCA) establishing models to analyze the value of this new method in clinical practice.
116 ischemic stroke patients and 26 normal people were enrolled. 58 patients were allocated into training dataset and 58 were divided into testing dataset along with 26 normal samples. The identification accuracy of our ResNet based approach in detecting the infarct core on non-contrast CT is 75.9%. The DCA shows that this deep learning method is capable of improving the net benefit of ischemic stroke patients.
Our deep learning residual network assisted with optimization methods is able to detect early infarct core on non-contrast CT images and has the potential to help physicians improve diagnostic accuracy in acute ischemic stroke patients.
探索一种主要基于深度学习残差网络（ResNet）的新方法，用于检测非增强 CT 图像上的梗死核心，提高急性缺血性脑卒中诊断的准确性。
本研究连续纳入磁共振弥散加权成像（MR-DWI）证实的首发缺血性脑卒中患者（发病时间：<9 h）和部分正常个体。所有患者均行 CT 平扫和 MR-DWI 扫描，扫描范围、层厚（4 mm）和层间距（4 mm）相同（两次检查的时间间隔：<4 h）。以 MR-DWI 作为梗死核心的金标准，采用深度学习 ResNet 结合最大后验概率（MAP）模型和后处理方法，检测非增强 CT 图像上的梗死核心。然后，我们使用决策曲线分析（DCA）建立模型，分析该新方法在临床实践中的价值。
共纳入 116 例缺血性脑卒中患者和 26 例正常个体。其中 58 例患者被分配到训练数据集，58 例患者和 26 例正常个体被分配到测试数据集。我们基于 ResNet 的方法在检测非增强 CT 上梗死核心的识别准确率为 75.9%。DCA 显示，这种深度学习方法能够提高缺血性脑卒中患者的净获益。
我们的深度学习残差网络辅助优化方法能够检测非增强 CT 图像上的早期梗死核心，有潜力帮助医生提高急性缺血性脑卒中患者的诊断准确性。
