利用深度学习确定胸部X光片上肺结核的活动情况
Deep Learning to Determine the Activity of Pulmonary Tuberculosis on Chest Radiographs.
作者信息
Lee Seowoo, Yim Jae-Joon, Kwak Nakwon, Lee Yeon Joo, Lee Jung-Kyu, Lee Ji Yeon, Kim Ju Sang, Kang Young Ae, Jeon Doosoo, Jang Myoung-Jin, Goo Jin Mo, Yoon Soon Ho
机构信息
From the Department of Radiology, Seoul National University College of Medicine, Seoul National University Hospital, 101 Daehak-ro, Chongno-gu, Seoul 03080, Korea (S.L., J.M.G., S.H.Y.); Division of Pulmonary and Critical Medicine, Department of Internal Medicine, Seoul National University College of Medicine, Seoul, Korea (J.J.Y., N.K.); Division of Pulmonary and Critical Care Medicine, Department of Internal Medicine, Seoul National University Bundang Hospital, Seongnam, Korea (Y.J.L.); Division of Pulmonary and Critical Care Medicine, Department of Internal Medicine, Seoul Metropolitan Government-Seoul National University Boramae Medical Center, Seoul, Korea (J.K.L.); Division of Pulmonary and Critical Care Medicine, Department of Internal Medicine, National Medical Center, Seoul, Korea (J.Y.L.); Division of Pulmonary and Critical Care Medicine, Department of Internal Medicine, Incheon St. Mary's Hospital, College of Medicine, The Catholic University of Korea, Incheon, Korea (J.S.K.); Division of Pulmonary and Critical Care Medicine, Department of Internal Medicine, Severance Hospital, Yonsei University College of Medicine, Seoul, Korea (Y.A.K.); Department of Internal Medicine, Pusan National University Yangsan Hospital, Pusan National University School of Medicine, Yangsan, Korea (D.J.); Medical Research Collaborating Center, Seoul National University Hospital, Seoul, Korea (M.J.J.); Institute of Radiation Medicine, Seoul National University Medical Research Center, Seoul, Korea (J.M.G.); and Department of Radiology, UMass Memorial Medical Center, Worcester, Mass (S.H.Y.).
出版信息
Radiology. 2021 Nov;301(2):435-442. doi: 10.1148/radiol.2021210063. Epub 2021 Aug 3.
Background Determining the activity of pulmonary tuberculosis on chest radiographs is difficult. Purpose To develop a deep learning model to identify active pulmonary tuberculosis on chest radiographs. Materials and Methods Chest radiographs were retrospectively gathered from a multicenter consecutive cohort with pulmonary tuberculosis who were successfully treated between 2011 and 2017, along with normal radiographs to enrich a negative class. The pretreatment and posttreatment radiographs were labeled as positive and negative classes, respectively. A neural network was trained with those radiographs to calculate the probability of active versus healed tuberculosis. A single-center consecutive cohort (test set 1; 89 patients, 148 radiographs) and data from one multicenter randomized controlled trial (test set 2; 366 patients, 3774 radiographs) were used to test the model. The area under the receiver operating characteristic curve (AUC) was used to evaluate the performance of the model and of the four expert readers. Results In total, 6654 pre- and posttreatment radiographs from 3327 patients (mean age ± standard deviation, 55 years ± 19; 1884 men) with pulmonary tuberculosis and 3182 normal radiographs from as many patients (mean age, 53 years ± 14; 1629 men) were gathered. For test set 1, the model showed a higher AUC (0.83; 95% CI: 0.73, 0.89) than one pulmonologist (0.69; 95% CI: 0.61, 0.76; < .001) and performed similarly to the other readers (AUC, 0.79-0.80; = .14-.23). For 200 randomly selected radiographs from test set 2, the model had a higher AUC (0.84) than the pulmonologists (0.71 and 0.74; < .001 and .01, respectively) and performed similarly to the radiologists (0.79 and 0.80; = .08 and .06, respectively). The model output increased by 0.30 on average with a higher degree of smear positivity (95% CI: 0.20, 0.39; < .001) and decreased during treatment (baseline, 3 months, and 6 months: 0.85, 0.51, and 0.26, respectively). Conclusion A deep learning model performed similarly to radiologists for accurately determining the activity of pulmonary tuberculosis on chest radiographs; it also was able to follow posttreatment changes. © RSNA, 2021
背景
在胸部X光片上确定肺结核的活动情况具有挑战性。目的：开发一种深度学习模型，用于在胸部X光片上识别活动性肺结核。材料与方法：回顾性收集了2011年至2017年期间成功治疗的多中心连续性肺结核患者队列的胸部X光片，以及正常X光片以丰富阴性类别。治疗前和治疗后的X光片分别标记为阳性和阴性类别。使用这些X光片训练神经网络，以计算活动性肺结核与已治愈肺结核的概率。使用单中心连续性队列（测试集1；89例患者，148张X光片）和一项多中心随机对照试验的数据（测试集2；366例患者，3774张X光片）来测试该模型。采用受试者操作特征曲线下面积（AUC）来评估模型和四位专家阅片者的表现。结果：共收集了3327例肺结核患者（平均年龄±标准差，55岁±19岁；1884例男性）的6654张治疗前和治疗后的X光片，以及同样数量患者（平均年龄，53岁±14岁；1629例男性）的3182张正常X光片。对于测试集1，该模型的AUC（0.83；95%CI：0.73，0.89）高于一位肺科医生（0.69；95%CI：0.61，0.76；P<0.001），与其他阅片者表现相似（AUC，0.79 - 0.80；P = 0.14 - 0.23）。对于从测试集2中随机选择的200张X光片，该模型的AUC（0.84）高于肺科医生（分别为0.71和0.74；P<0.001和P<0.01），与放射科医生表现相似（分别为0.79和0.80；P = 0.08和P = 0.06）。随着涂片阳性程度越高，模型输出平均增加0.30（95%CI：0.20，0.39；P<0.001），且在治疗期间下降（基线、3个月和6个月时分别为0.85、0.51和0.26）。结论：一种深度学习模型在通过胸部X光片准确确定肺结核活动情况方面与放射科医生表现相似；它还能够跟踪治疗后的变化。©RSNA，2021
Zotero 插件