Lu Mingming, Zheng Yijia, Liu Shitong, Zhang Xiaolan, Lv Jiahui, Liu Yuan, Li Baobao, Yuan Fei, Peng Peng, Han Cong, Ma Chune, Zheng Chao, Zhang Hongtao, Cai Jianming
Department of Radiology, Pingjin Hospital, Characteristic Medical Center of Chinese People's Armed Police Force, Tianjin, China.
Department of Radiology, The Fifth Medical Center, Chinese PLA General Hospital, Beijing, China.
EClinicalMedicine. 2024 Nov 5;77:102888. doi: 10.1016/j.eclinm.2024.102888. eCollection 2024 Nov.
This study explores the potential of the deep learning-based convolutional neural network (CNN) to automatically recognize MMD using MRA images from atherosclerotic disease (ASD) and normal control (NC).
In this retrospective study in China, 600 participants (200 MMD, 200 ASD and 200 NC) were collected from one institution as an internal dataset for training and 60 from another institution were collected as external testing set for validation. All participants were divided into training (N = 450) and validation sets (N = 90), internal testing set (N = 60), and external testing set (N = 60). The input to the CNN models comprised preprocessed MRA images, while the output was a tripartite classification label that identified the patient's diagnostic group. The performances of 3D CNN models were evaluated using a comprehensive set of metrics such as area under the curve (AUC) and accuracy. Gradient-weighted Class Activation Mapping (Grad-CAM) was used to visualize the CNN's decision-making process in MMD diagnosis by highlighting key areas. Finally, the diagnostic performances of the CNN models were compared with those of two experienced radiologists.
DenseNet-121 exhibited superior discrimination capabilities, achieving a macro-average AUC of 0.977 (95% CI, 0.928-0.995) in the internal test sets and 0.880 (95% CI, 0.786-0.937) in the external validation sets, thus exhibiting comparable diagnostic capabilities to those of human radiologists. In the binary classification where ASD and NC were group together, with MMD as the separate group for targeted detection, DenseNet-121 achieved an accuracy of 0.967 (95% CI, 0.886-0.991). Additionally, the Grad-CAM results for the MMD, with areas of intense redness indicating critical areas identified by the model, reflected decision-making similar to human experts.
This study highlights the efficacy of CNN model in the automated diagnosis of MMD on MRA images, easing the workload on radiologists and promising integration into clinical workflows.
National Natural Science Foundation of China, Tianjin Science and Technology Project and Beijing Natural Science Foundation.
本研究探讨基于深度学习的卷积神经网络（CNN）利用动脉粥样硬化疾病（ASD）和正常对照（NC）的磁共振血管造影（MRA）图像自动识别烟雾病（MMD）的潜力。
在中国进行的这项回顾性研究中，从一个机构收集了600名参与者（200名MMD患者、200名ASD患者和200名NC）作为内部训练数据集，从另一个机构收集了60名参与者作为外部测试集用于验证。所有参与者被分为训练集（N = 450）和验证集（N = 90）、内部测试集（N = 60）和外部测试集（N = 60）。CNN模型的输入包括预处理后的MRA图像，而输出是一个三方分类标签，用于识别患者的诊断组。使用一组综合指标（如曲线下面积（AUC）和准确率）评估3D CNN模型的性能。通过突出关键区域，使用梯度加权类激活映射（Grad-CAM）来可视化CNN在MMD诊断中的决策过程。最后，将CNN模型的诊断性能与两位经验丰富的放射科医生的诊断性能进行比较。
DenseNet-121表现出卓越的辨别能力，在内部测试集中实现了0.977（95%CI，0.928 - 0.995）的宏观平均AUC，在外部验证集中实现了0.880（95%CI，0.786 - 0.937）的宏观平均AUC，因此展现出与人类放射科医生相当的诊断能力。在将ASD和NC归为一组、MMD作为单独目标检测组的二元分类中，DenseNet-121的准确率达到了0.967（95%CI，0.886 - 0.991）。此外，MMD的Grad-CAM结果中，深红色区域表明是模型识别出的关键区域，反映出与人类专家相似的决策过程。
本研究突出了CNN模型在基于MRA图像自动诊断MMD方面的有效性，减轻了放射科医生的工作量，并有望整合到临床工作流程中。
中国国家自然科学基金、天津市科技项目和北京市自然科学基金。
