Segmentation of Lung Nodules on CT Images Using a Nested Three-Dimensional Fully Connected Convolutional Network

Shoji Kido; Shunske Kidera; Yasushi Hirano; Shingo Mabu; Tohru Kamiya; Nobuyuki Tanaka; Yuki Suzuki; Masahiro Yanagawa; Noriyuki Tomiyama

doi:10.3389/frai.2022.782225

Segmentation of Lung Nodules on CT Images Using a Nested Three-Dimensional Fully Connected Convolutional Network

Front Artif Intell. 2022 Feb 17:5:782225. doi: 10.3389/frai.2022.782225. eCollection 2022.

Authors

Shoji Kido¹, Shunske Kidera², Yasushi Hirano³, Shingo Mabu², Tohru Kamiya⁴, Nobuyuki Tanaka⁵, Yuki Suzuki¹, Masahiro Yanagawa⁶, Noriyuki Tomiyama⁶

Affiliations

¹ Department of Artificial Intelligence Diagnostic Radiology, Osaka University Graduate School of Medicine, Suita, Japan.
² Graduate School of Sciences and Technology for Innovation, Yamaguchi University, Ube, Japan.
³ Medical Informatics and Decision Sciences, Yamaguchi University Hospital, Ube, Japan.
⁴ Department of Mechanical and Control Engineering, Faculty of Engineering Kyushu Institute of Technology, Kitakyushu, Japan.
⁵ Department of Radiology, National Hospital Organization, Yamaguchi-Ube Medical Center, Ube, Japan.
⁶ Department of Radiology, Osaka University Graduate School of Medicine, Suita, Japan.

Abstract

In computer-aided diagnosis systems for lung cancer, segmentation of lung nodules is important for analyzing image features of lung nodules on computed tomography (CT) images and distinguishing malignant nodules from benign ones. However, it is difficult to accurately and robustly segment lung nodules attached to the chest wall or with ground-glass opacities using conventional image processing methods. Therefore, this study aimed to develop a method for robust and accurate three-dimensional (3D) segmentation of lung nodule regions using deep learning. In this study, a nested 3D fully connected convolutional network with residual unit structures was proposed, and designed a new loss function. Compared with annotated images obtained under the guidance of a radiologist, the Dice similarity coefficient (DS) and intersection over union (IoU) were 0.845 ± 0.008 and 0.738 ± 0.011, respectively, for 332 lung nodules (lung adenocarcinoma) obtained from 332 patients. On the other hand, for 3D U-Net and 3D SegNet, the DS was 0.822 ± 0.009 and 0.786 ± 0.011, respectively, and the IoU was 0.711 ± 0.011 and 0.660 ± 0.012, respectively. These results indicate that the proposed method is significantly superior to well-known deep learning models. Moreover, we compared the results obtained from the proposed method with those obtained from conventional image processing methods, watersheds, and graph cuts. The DS and IoU results for the watershed method were 0.628 ± 0.027 and 0.494 ± 0.025, respectively, and those for the graph cut method were 0.566 ± 0.025 and 0.414 ± 0.021, respectively. These results indicate that the proposed method is significantly superior to conventional image processing methods. The proposed method may be useful for accurate and robust segmentation of lung nodules to assist radiologists in the diagnosis of lung nodules such as lung adenocarcinoma on CT images.

Keywords: SegNet; U-Net; computer-aided diagnosis; deep learning; graph cut; lung nodule; segmentation; watershed.