From the Department of Radiology and Nuclear Medicine, Xuanwu Hospital, Capital Medical University, No. 45 Changchun St, Xicheng District, Beijing 100053, China (F.F., Y.S., M.Z., D.R., J.L.); Beijing Key Laboratory of Magnetic Resonance Imaging and Brain Informatics, Beijing, China (F.F., Y.S., M.Z., D.R., J.L.); Department of Nuclear Medicine, Ruijin Hospital, Shanghai Jiaotong University, Shanghai, China (F.F.); Shukun (Beijing) Technology Co, Beijing, China (G.Y., C.Z.); and Department of Radiology, Shandong Provincial Hospital, Jinan, China (X.W.).
Radiology. 2023 May;307(3):e220996. doi: 10.1148/radiol.220996. Epub 2023 Mar 7.
Background Studies have rarely investigated stenosis detection from head and neck CT angiography scans because accurate interpretation is time consuming and labor intensive. Purpose To develop an automated convolutional neural network-based method for accurate stenosis detection and plaque classification in head and neck CT angiography images and compare its performance with that of radiologists. Materials and Methods A deep learning (DL) algorithm was constructed and trained with use of head and neck CT angiography images that were collected retrospectively from four tertiary hospitals between March 2020 and July 2021. CT scans were partitioned into training, validation, and independent test sets at a ratio of 7:2:1. An independent test set of CT angiography scans was collected prospectively between October 2021 and December 2021 in one of the four tertiary centers. Stenosis grade categories were as follows: mild stenosis (<50%), moderate stenosis (50%-69%), severe stenosis (70%-99%), and occlusion (100%). The stenosis diagnosis and plaque classification of the algorithm were compared with the ground truth of consensus by two radiologists (with more than 10 years of experience). The performance of the models was analyzed in terms of accuracy, sensitivity, specificity, and areas under the receiver operating characteristic curve. Results There were 3266 patients (mean age ± SD, 62 years ± 12; 2096 men) evaluated. The consistency between radiologists and the DL-assisted algorithm on plaque classification was 85.6% (320 of 374 cases [95% CI: 83.2, 88.6]) on a per-vessel basis. Moreover, the artificial intelligence model assisted in visual assessment, such as increasing confidence in the degree of stenosis. This reduced the time needed for diagnosis and report writing of radiologists from 28.8 minutes ± 5.6 to 12.4 minutes ± 2.0 ( < .001). Conclusion A deep learning algorithm for head and neck CT angiography interpretation accurately determined vessel stenosis and plaque classification and had equivalent diagnostic performance when compared with experienced radiologists. © RSNA, 2023
背景 研究很少涉及从头颈部 CT 血管造影扫描中检测狭窄，因为准确的解释既费时又费力。 目的 开发一种基于卷积神经网络的自动方法，对头颈部 CT 血管造影图像进行准确的狭窄检测和斑块分类，并将其性能与放射科医生进行比较。 材料与方法 使用回顾性收集的来自 4 家三级医院的 2020 年 3 月至 2021 年 7 月期间的头颈部 CT 血管造影图像构建和训练深度学习（DL）算法。CT 扫描按 7:2:1 的比例分为训练集、验证集和独立测试集。2021 年 10 月至 12 月期间，在其中一家 4 家三级医院前瞻性收集了独立的 CT 血管造影扫描的测试集。狭窄等级类别如下：轻度狭窄（<50%）、中度狭窄（50%-69%）、重度狭窄（70%-99%）和闭塞（100%）。该算法的狭窄诊断和斑块分类与两名具有 10 年以上经验的放射科医生的共识进行了比较。通过准确性、敏感性、特异性和受试者工作特征曲线下面积分析模型的性能。 结果 共评估了 3266 例患者（平均年龄±标准差，62 岁±12 岁；2096 例男性）。在每支血管的基础上，放射科医生和基于深度学习的辅助算法在斑块分类方面的一致性为 85.6%（320 例中有 374 例[95%CI：83.2，88.6]）。此外，人工智能模型有助于进行视觉评估，例如增加对狭窄程度的信心。这使得放射科医生的诊断和报告撰写时间从 28.8 分钟±5.6 减少到 12.4 分钟±2.0（<0.001）。 结论 用于头颈部 CT 血管造影解读的深度学习算法可以准确确定血管狭窄和斑块分类，与经验丰富的放射科医生相比具有相当的诊断性能。 © RSNA，2023
