Automatic lung nodule detection using a 3D deep convolutional neural network combined with a multi-scale prediction strategy in chest CTs. (1st December 2018)

• Record Type: Journal Article
• Authors: Gu, Yu; Lu, Xiaoqi; Yang, Lidong; Zhang, Baohua; Yu, Dahua; Zhao, Ying; Gao, Lixin; Wu, Liang; Zhou, Tao
• Is Part Of: Computers in biology and medicine. Volume 103(2018)
• Page Start: 220
• Page End: 231
• Publication Date: 2018-12-01
• Subjects: Lung nodule detection -- 3D convolutional neural network -- Multi-scale cube prediction -- Cube clustering -- Deep learning; Medicine -- Data processing -- Periodicals; Biology -- Data processing -- Periodicals; 610.285
• DOI: 10.1016/j.compbiomed.2018.10.011
• Languages: English
• ISSNs: 0010-4825

Abstract

Abstract: Objective: A novel computer-aided detection (CAD) scheme for lung nodule detection using a 3D deep convolutional neural network combined with a multi-scale prediction strategy is proposed to assist radiologists by providing a second opinion on accurate lung nodule detection, which is a crucial step in early diagnosis of lung cancer. Method: A 3D deep convolutional neural network (CNN) with multi-scale prediction was used to detect lung nodules after the lungs were segmented from chest CT scans, with a comprehensive method utilized. Compared with a 2D CNN, a 3D CNN can utilize richer spatial 3D contextual information and generate more discriminative features after being trained with 3D samples to fully represent lung nodules. Furthermore, a multi-scale lung nodule prediction strategy, including multi-scale cube prediction and cube clustering, is also proposed to detect extremely small nodules. Result: The proposed method was evaluated on 888 thin-slice scans with 1186 nodules in the LUNA16 database. All results were obtained via 10-fold cross-validation. Three options of the proposed scheme are provided for selection according to the actual needs. The sensitivity of the proposed scheme with the primary option reached 87.94% and 92.93% at one and four false positives per scan, respectively. Meanwhile, the competition performance metric (CPM) score is very satisfying (0.7967). Conclusion: The experimental results demonstrate the outstanding detection performance of the proposed nodule detection scheme. In addition, the proposed scheme can be extended to other medical image recognition fields. Highlights: A 3D CNN with satisfactory performance and multi-scale prediction is proposed. Augmentation of test data was performed to detect small nodules. Cubes with small scales are particularly suitable for detecting small nodules. Cubes are clustered with the DBSCAN algorithm to improve performance. Three options of the proposed scheme are provided for selection according to the actual needs.