Department of Radiology, Seoul St. Mary's Hospital, College of Medicine, The Catholic University of Korea, Seoul, Republic of Korea; Department of Radiology, Center for Clinical Imaging Data Science, Research Institute of Radiological Sciences, Severance Hospital, Yonsei University College of Medicine, Seoul, Republic of Korea.
Department of Radiology, Center for Clinical Imaging Data Science, Research Institute of Radiological Sciences, Severance Hospital, Yonsei University College of Medicine, Seoul, Republic of Korea; Department of Biomedical Systems Informatics, Yonsei University College of Medicine, Seoul, Republic of Korea.
Eur J Radiol. 2021 Apr;137:109582. doi: 10.1016/j.ejrad.2021.109582. Epub 2021 Feb 6.
We aimed to develop a deep learning (DL)-based algorithm for automated quantification of aortic valve calcium (AVC) from non-enhanced electrocardiogram-gated cardiac CT scans and compare performance of DL-measured AVC volume and Agatston score with those of visual gradings by radiologist readers for classification of AVC severity.
A total of 589 CT examinations performed at a single center between March 2010 and August 2017 were retrospectively included. The DL algorithm was designed to segment AVC and to quantify AVC volume, and Agatston score was calculated using attenuation values. Manually measured AVC volume and Agatston score were used as ground truth. To validate AVC segmentation performance, the Dice coefficient was calculated. For observer performance testing, four radiologists determined AVC grade in two reading rounds. The diagnostic performance of DL-measured AVC volume and Agaston score for classifying severe AVC was compared with that of each reader's assessment.
After applying the DL algorithm, the Dice coefficient score was 0.807. In patients with AVC, accuracy of DL-measured AVC volume for AVC grading was 97.0 % with area under the curve (AUC) of 0.964 (95 % confidence interval [CI] 0.923-1) in the test set, which was better than the radiologist readers (accuracy 69.7 %-91.9 %, AUC 0.762-0.923) with manually measured AVC volume as ground truth. When manually measured AVC Agatston score was used as ground truth, accuracy of DL-measured AVC Agatston score for AVC grading was 92.9 % with AUC of 0.933 (95 % CI 0.885-0.981) in the test set, which was also better than the radiologist readers (accuracy 77.8-89.9 %, AUC 0.791-0.903).
DL-based automated AVC quantification may be comparable with manual measurements. The diagnostic performance of the DL-measured AVC volume and Agatston score for classification of severe AVC outperforms radiologist readers.
我们旨在开发一种基于深度学习（DL）的算法，用于从非增强心电图门控心脏 CT 扫描中自动量化主动脉瓣钙（AVC），并比较 DL 测量的 AVC 体积和 Agatston 评分与放射科医师读者对 AVC 严重程度进行分类的视觉分级的性能。
回顾性纳入 2010 年 3 月至 2017 年 8 月在一家中心进行的总共 589 次 CT 检查。该 DL 算法旨在分割 AVC 并量化 AVC 体积，并使用衰减值计算 Agatston 评分。手动测量的 AVC 体积和 Agatston 评分被用作基准。为了验证 AVC 分割性能，计算了 Dice 系数。对于观察者性能测试，四位放射科医师在两轮阅读中确定 AVC 等级。比较了 DL 测量的 AVC 体积和 Agaston 评分对分类严重 AVC 的诊断性能与每位读者评估的诊断性能。
应用 DL 算法后，Dice 系数评分为 0.807。在患有 AVC 的患者中，对于 AVC 分级，DL 测量的 AVC 体积的准确性为 97.0%，测试集的曲线下面积（AUC）为 0.964（95%置信区间[CI] 0.923-1），优于以手动测量的 AVC 体积作为基准的放射科医师读者（准确性 69.7%-91.9%，AUC 0.762-0.923）。当以手动测量的 AVC Agatston 评分作为基准时，对于 AVC 分级，DL 测量的 AVC Agatston 评分的准确性为 92.9%，测试集的 AUC 为 0.933（95%CI 0.885-0.981），也优于放射科医师读者（准确性 77.8%-89.9%，AUC 0.791-0.903）。
基于 DL 的自动 AVC 量化可能与手动测量相当。DL 测量的 AVC 体积和 Agatston 评分对严重 AVC 分类的诊断性能优于放射科医师读者。
