Automated Pulmonary Nodule Classification in Computed Tomography Images Using a Deep Convolutional Neural Network Trained by Generative Adversarial Networks

被引:71
作者
Onishi, Yuya [1 ]
Teramoto, Atsushi [1 ]
Tsujimoto, Masakazu [2 ]
Tsukamoto, Tetsuya [3 ]
Saito, Kuniaki [1 ]
Toyama, Hiroshi [3 ]
Imaizumi, Kazuyoshi [3 ]
Fujita, Hiroshi [4 ]
机构
[1] Fujita Hlth Univ, Grad Sch Hlth Sci, 1-98 Dengakugakubo,Kutsukake Cho, Toyoake, Aichi 4701192, Japan
[2] Fujita Hlth Univ Hosp, 1-98 Dengakugakubo,Kutsukake Cho, Toyoake, Aichi 4701192, Japan
[3] Fujita Hlth Univ, Sch Med, 1-98 Dengakugakubo,Kutsukake Cho, Toyoake, Aichi 4701192, Japan
[4] Gifu Univ, 1-1 Yanagido, Gifu 5011194, Japan
关键词
LUNG-CANCER;
D O I
10.1155/2019/6051939
中图分类号
Q81 [生物工程学(生物技术)]; Q93 [微生物学];
学科分类号
071005 [微生物学]; 090105 [作物生产系统与生态工程];
摘要
Lung cancer is a leading cause of death worldwide. Although computed tomography (CT) examinations are frequently used for lung cancer diagnosis, it can be difficult to distinguish between benign and malignant pulmonary nodules on the basis of CT images alone. Therefore, a bronchoscopic biopsy may be conducted if malignancy is suspected following CT examinations. However, biopsies are highly invasive, and patients with benign nodules may undergo many unnecessary biopsies. To prevent this, an imaging diagnosis with high classification accuracy is essential. In this study, we investigate the automated classification of pulmonary nodules in CT images using a deep convolutional neural network (DCNN). We use generative adversarial networks (GANs) to generate additional images when only small amounts of data are available, which is a common problem in medical research, and evaluate whether the classification accuracy is improved by generating a large amount of new pulmonary nodule images using the GAN. Using the proposed method, CT images of 60 cases with confirmed pathological diagnosis by biopsy are analyzed. The benign nodules assessed in this study are difficult for radiologists to differentiate because they cannot be rejected as being malignant. A volume of interest centered on the pulmonary nodule is extracted from the CT images, and further images are created using axial sections and augmented data. The DCNN is trained using nodule images generated by the GAN and then fine-tuned using the actual nodule images to allow the DCNN to distinguish between benign and malignant nodules. This pretraining and fine-tuning process makes it possible to distinguish 66.7% of benign nodules and 93.9% of malignant nodules. These results indicate that the proposed method improves the classification accuracy by approximately 20% in comparison with training using only the original images.
引用
收藏
页数:9
相关论文
共 36 条
[1]
[Anonymous], 2018, CANCER FACTS AND FIG
[2]
[Anonymous], 2016, P 30 INT C NEURAL IN
[3]
[Anonymous], MED IMAGE SYNTHESIS
[4]
[Anonymous], PROC CVPR IEEE
[5]
[Anonymous], 2015, ARXIV PREPRINT ARXIV
[6]
[Anonymous], PROGRESSIVE GROWING
[7]
[Anonymous], 2015, DEEP LEARNING NATURE, DOI [10.1038/nature14539, DOI 10.1038/NATURE14539]
[8]
[Anonymous], FOOL RADIOLOGISTS GE
[9]
[Anonymous], 2017, ADVERSARIAL RETINAL
[10]
[Anonymous], LOW DOSE CT IMAGE DE