From the Department of Radiology, Fujita Health University School of Medicine, 1-98 Dengakugakubo, Kutsukake-cho, Toyoake 470-1192, Japan (Y.O.); Joint Research Laboratory of Advanced Medical Imaging, Fujita Health University School of Medicine, Toyoake, Japan (Y.O.); Division of Functional and Diagnostic Imaging Research, Department of Radiology, Kobe University Graduate School of Medicine, Kobe, Japan (Y.O., S.S., T.Y.); Canon Medical Systems, Otawara, Japan (K.A.); Corporate Research and Development Center, Toshiba, Kawasaki, Japan (A.Y.); Division of Radiology, Department of Radiology, Kobe University Graduate School of Medicine, Kobe, Japan (Y.U., Y.K.); Department of Radiology, Kohnan Hospital, Kobe, Japan (Y.K.); and Department of Radiology, Hyogo Cancer Center, Akashi, Japan (D.T.).
Radiology. 2020 Aug;296(2):432-443. doi: 10.1148/radiol.2020191740. Epub 2020 May 26.
Background Deep learning may help to improve computer-aided detection of volume (CADv) measurement of pulmonary nodules at chest CT. Purpose To determine the efficacy of a deep learning method for improving CADv for measuring the solid and ground-glass opacity (GGO) volumes of a nodule, doubling time (DT), and the change in volume at chest CT. Materials and Methods From January 2014 to December 2016, patients with pulmonary nodules at CT were retrospectively reviewed. CADv without and with a convolutional neural network (CNN) automatically determined total nodule volume change per day and DT. Area under the curves (AUCs) on a per-nodule basis and diagnostic accuracy on a per-patient basis were compared among all indexes from CADv with and without CNN for differentiating benign from malignant nodules. Results The CNN training set was 294 nodules in 217 patients, the validation set was 41 nodules in 32 validation patients, and the test set was 290 nodules in 188 patients. A total of 170 patients had 290 nodules (mean size ± standard deviation, 11 mm ± 5; range, 4-29 mm) diagnosed as 132 malignant nodules and 158 benign nodules. There were 132 solid nodules (46%), 106 part-solid nodules (36%), and 52 ground-glass nodules (18%). The test set results showed that the diagnostic performance of the CNN with CADv for total nodule volume change per day was larger than DT of CADv with CNN (AUC, 0.94 [95% confidence interval {CI}: 0.90, 0.96] vs 0.67 [95% CI: 0.60, 0.74]; < .001) and CADv without CNN (total nodule volume change per day: AUC, 0.69 [95% CI: 0.62, 0.75]; < .001; DT: AUC, 0.58 [95% CI: 0.51, 0.65]; < .001). The accuracy of total nodule volume change per day of CADv with CNN was significantly higher than that of CADv without CNN ( < .001) and DT of both methods ( < .001). Conclusion Convolutional neural network is useful for improving accuracy of computer-aided detection of volume measurement and nodule differentiation capability at CT for patients with pulmonary nodules. © RSNA, 2020
背景 深度学习有助于提高胸部 CT 中肺结节容积计算机辅助检测（CADv）的测量水平。目的 旨在评估一种深度学习方法对提高 CADv 测量结节实性和磨玻璃密度（GGO）容积、倍增时间（DT）和 CT 上容积变化的效能。材料与方法 本研究回顾性分析了 2014 年 1 月至 2016 年 12 月期间 CT 检出的肺结节患者。CADv 结合和不结合卷积神经网络（CNN）自动确定了每天结节总容积的变化量和 DT。比较了 CADv 结合和不结合 CNN 的所有指标在基于结节和基于患者的基础上区分良恶性结节的曲线下面积（AUC）和诊断准确性。结果 CNN 训练集为 217 例患者的 294 个结节，验证集为 32 例验证患者的 41 个结节，测试集为 188 例患者的 290 个结节。共 170 例患者的 290 个结节（平均大小±标准差，11mm±5；范围，4-29mm）被诊断为 132 个恶性结节和 158 个良性结节。其中实性结节 132 个（46%），部分实性结节 106 个（36%），磨玻璃结节 52 个（18%）。测试集结果显示，CADv 结合 CNN 的结节总容积日变化量的诊断性能大于 CADv 结合 CNN 的 DT（AUC：0.94[95%CI：0.90，0.96]比 0.67[95%CI：0.60，0.74]；<0.001）和 CADv 不结合 CNN 的总容积日变化量（AUC：0.69[95%CI：0.62，0.75]；<0.001；DT：AUC：0.58[95%CI：0.51，0.65]；<0.001）。CADv 结合 CNN 的总容积日变化量的准确性明显高于 CADv 不结合 CNN（<0.001）和两种方法的 DT（<0.001）。结论 卷积神经网络有助于提高 CT 对肺结节的容积测量和结节鉴别能力，提高计算机辅助检测的准确性。 © 2020 RSNA
