3D deep learning for detecting pulmonary nodules in CT scans
Open Access
- 22 August 2018
- journal article
- research article
- Published by Oxford University Press (OUP) in Journal of the American Medical Informatics Association
- Vol. 25 (10), 1301-1310
- https://doi.org/10.1093/jamia/ocy098
Abstract
Objective: To demonstrate and test the validity of a novel deep-learning-based system for the automated detection of pulmonary nodules.Materials and Methods: The proposed system uses 2 3D deep learning models, 1 for each of the essential tasks of computer-aided nodule detection: candidate generation and false positive reduction. A total of 888 scans from the LIDC-IDRI dataset were used for training and evaluation.Results: Results for candidate generation on the test data indicated a detection rate of 94.77% with 30.39 false positives per scan, while the test results for false positive reduction exhibited a sensitivity of 94.21% with 1.789 false positives per scan. The overall system detection rate on the test data was 89.29% with 1.789 false positives per scan.Discussion: An extensive and rigorous validation was conducted to assess the performance of the proposed system. The system demonstrated a novel combination of 3D deep neural network architectures and demonstrates the use of deep learning for both candidate generation and false positive reduction to be evaluated with a substantial test dataset. The results strongly support the ability of deep learning pulmonary nodule detection systems to generalize to unseen data. The source code and trained model weights have been made available.Conclusion: A novel deep-neural-network-based pulmonary nodule detection system is demonstrated and validated. The results provide comparison of the proposed deep-learning-based system over other similar systems based on performance.Keywords
Funding Information
- Auburn University
- National Cancer Institute and the Foundation
- National Institutes of Health
This publication has 53 references indexed in Scilit:
- Automated Pulmonary Nodule Detection System in Computed Tomography Images: A Hierarchical Block Classification ApproachEntropy, 2013
- Reduced Lung-Cancer Mortality with Low-Dose Computed Tomographic ScreeningThe New England Journal of Medicine, 2011
- The National Lung Screening Trial: Overview and Study DesignRadiology, 2011
- A new computationally efficient CAD system for pulmonary nodule detection in CT imageryMedical Image Analysis, 2010
- Area under the Free‐Response ROC Curve (FROC) and a Related Summary IndexBiometrics, 2009
- Errors in Cancer Diagnosis: Current Understanding and Future DirectionsJournal of Clinical Oncology, 2007
- Feature Subset Selection for Improving the Performance of False Positive Reduction in Lung Nodule CADIEEE Transactions on Information Technology in Biomedicine, 2006
- Gradient-based learning applied to document recognitionProceedings of the IEEE, 1998
- Artificial convolution neural network techniques and applications for lung nodule detectionIEEE Transactions on Medical Imaging, 1995
- Measures of the Amount of Ecologic Association Between SpeciesEcology, 1945