Chen, C.-H.C.-H.ChenLee, Y.-W.Y.-W.LeeHuang, Y.-S.Y.-S.HuangLan, W.-R.W.-R.LanChang, R.-F.R.-F.ChangTu, C.-Y.C.-Y.TuChen, C.-Y.C.-Y.ChenLiao, W.-C.W.-C.LiaoRUEY-FENG CHANG2020-05-042020-05-042019https://scholars.lib.ntu.edu.tw/handle/123456789/489567Background and objective: In the United States, lung cancer is the leading cause of cancer death. The survival rate could increase by early detection. In recent years, the endobronchial ultrasonography (EBUS)images have been utilized to differentiate between benign and malignant lesions and guide transbronchial needle aspiration because it is real-time, radiation-free and has better performance. However, the diagnosis depends on the subjective judgment from doctors. In some previous studies, which using the grayscale image textures of the EBUS images to classify the lung lesions but it belonged to semi-automated system which still need the experts to select a part of the lesion first. Therefore, the main purpose of this study was to achieve full automation assistance by using convolution neural network. Methods: First of all, the EBUS images resized to the input size of convolution neural network (CNN). And then, the training data were rotated and flipped. The parameters of the model trained with ImageNet previously were transferred to the CaffeNet used to classify the lung lesions. And then, the parameter of the CaffeNet was optimized by the EBUS training data. The features with 4096 dimension were extracted from the 7th fully connected layer and the support vector machine (SVM)was utilized to differentiate benign and malignant. This study was validated with 164 cases including 56 benign and 108 malignant. Results: According to the experiment results, applying the classification by the features from the CNN with transfer learning had better performance than the conventional method with gray level co-occurrence matrix (GLCM)features. The accuracy, sensitivity, specificity, and the area under ROC achieved 85.4% (140/164), 87.0% (94/108), 82.1% (46/56), and 0.8705, respectively. Conclusions: From the experiment results, it has potential ability to diagnose EBUS images with CNN. ? 2019 Elsevier B.V.Computer-aided diagnosis (CAD); Convolutional neural network (CNN); Endobronchial ultrasound images (EBUS); Lung cancer; Transfer learning[SDGs]SDG3Automation; Biological organs; Computer aided instruction; Computer networks; Convolution; Diseases; Image texture; Neural networks; Support vector machines; Textures; Transfer matrix method; Ultrasonic applications; Computer Aided Diagnosis(CAD); Convolutional neural network; Lung Cancer; Transfer learning; Ultrasound images; Computer aided diagnosis; adult; aged; Article; automation; cancer diagnosis; computer assisted diagnosis; computer language; controlled study; convolution neural network; diagnostic accuracy; endobronchial ultrasonography; feature extraction; human; lung cancer; machine learning; major clinical study; predictive value; sensitivity and specificity; support vector machine; transfer of learning; algorithm; area under the curve; automated pattern recognition; bronchus; diagnostic imaging; echography; lung; lung tumor; middle aged; receiver operating characteristic; reproducibility; very elderly; young adult; Adult; Aged; Aged, 80 and over; Algorithms; Area Under Curve; Bronchi; Diagnosis, Computer-Assisted; Humans; Lung; Lung Neoplasms; Middle Aged; Neural Networks, Computer; Pattern Recognition, Automated; Reproducibility of Results; ROC Curve; Sensitivity and Specificity; Ultrasonography; Young Adultjournal article10.1016/j.cmpb.2019.05.020https://www.scopus.com/inward/record.uri?eid=2-s2.0-85066448694&doi=10.1016%2fj.cmpb.2019.05.020&partnerID=40&md5=4636dcc8a41deb736c29d44a6b5ffcb4