ABSTRACT
Mediastinal lymph nodes are one of the most critical factors to identify the clinical stages of lung cancer. As the lymph nodes are low in attenuation and cluttering with various shapes and sizes, manual detection is usually error-prone and effort-intensive. This paper introduces a method for automatic detection of mediastinal lymph nodes by proposing three significant contributions. First, we constraint the detection area, mediastinal region, using grey-level thresholding. Next, we apply the watershed method and hessian eigenvalues to separate a cluster of lymph nodes. Finally, we build a three-dimensional convolutional neural network (3D CNN) to distinguish the actual lymph nodes from other false lesions. Our experiment is conducted using 70 CT exams containing 314 lymph nodes and achieved a favorable result with 94 % detection rate.
- International Agency for Research on Cancer, Latest global cancer data: Cancer burden rises to 18.1 million new cases and 9.6 million cancer deaths in 2018, World Health Organization. September 2018.Google Scholar
- Feuerstein M, Deguchi D, Kitasaka T, et al. Automatic mediastinal lymph node detection in chest CT. In: Karssemeijer N, Giger ML, eds. Lake Buena Vista, FL; 2009:72600V. doi:10.1117/12.811101Google Scholar
- Feulner J, Kevin Zhou S, Hammon M, Hornegger J, Comaniciu D. Lymph node detection and segmentation in chest CT data using discriminative learning and a spatial prior. Medical Image Analysis. 2013;17(2):254--270. doi:10.1016/j.media.2012.11.001Google Scholar
- Liu J, Zhao J, Hoffman J, et al. Detection and station mapping of mediastinal lymph nodes on thoracic computed tomography using spatial prior from multi-atlas label fusion. In: 2014 IEEE 11th International Symposium on Biomedical Imaging (ISBI). Beijing, China: IEEE; 2014:1107--1110. doi:10.1109/ISBI.2014.6868068Google ScholarCross Ref
- Lai H, Chen W, Fu H. A new double-sampling method for mediastinal lymph nodes detection by deep conventional neural network. In: 2018 Chinese Control And Decision Conference (CCDC). Shenyang, China: IEEE; 2018:6286--6290. doi:10.1109/CCDC.2018.8408233Google Scholar
- Holger, Roth, Lu, Le, Seff, Ari, Cherry, Kevin M, Hoffman, Joanne, Wang, Shijun, ... Summers, Ronald M. A new 2.5 D representation for lymph node detection in CT. The Cancer Imaging Archive. 2015. http://doi.org/10.7937/K9/TCIA.2015.AQIIDCNM.Google Scholar
- Clark K, Vendt B, Smith K, et al. The Cancer Imaging Archive (TCIA): Maintaining and Operating a Public Information Repository. Journal of Digital Imaging. 2013;26(6):1045--1057. doi:10.1007/s10278-013-9622-7Google Scholar
- Beucher S, Lantuéjoul C. Use of Watersheds in Contour Detection.:13.Google Scholar
- Jerman T, Pernuš F, Likar B, Špiclin Ž. Beyond Frangi: an improved multiscale vesselness filter. In: Ourselin S, Styner MA, eds. Orlando, Florida, United States; 2015:94132A. doi:10.1117/12.2081147Google Scholar
- Liu J, Hoffman J, Zhao J, et al. Mediastinal lymph node detection and station mapping on chest CT using spatial priors and random forest. Medical Physics. 2016;43(7):4362--4374. doi:10.1118/1.4954009Google Scholar
- Allaouzi I, Benamrou B, Benamrou M, Ahmed MB. 3D-CNN and SVM for Multi-Drug Resistance Detection.Google Scholar
Index Terms
- Automatic Detection of Mediastinal Lymph Nodes using 3D Convolutional Neural Network
Recommendations
Lung-Nodule Segmentation Using a Convolutional Neural Network with the U-Net Architecture
Pattern RecognitionAbstractLung cancer is one of the types of cancer that claims the most lives globally. For screening purposes, computed tomography scans are the most reliable source for nodule detection, as it reveals the structure of the chest, through a three ...
Automated CAD for detection of lung nodule using CT scans
COMPUTE '10: Proceedings of the Third Annual ACM Bangalore ConferenceThe main objective of this paper is to evaluate the performance of the Computer-Aided Detection (CAD) system for automated nodule detection in lungs using CT scan images. The CAD system is applied to CT scans collected in a screening program for lung ...
The Effect of Artificial Neural Network Model Combined with Six Tumor Markers in Auxiliary Diagnosis of Lung Cancer
To evaluate the diagnosis potential of artificial neural network (ANN) model combined with six tumor markers in auxiliary diagnosis of lung cancer, to differentiate lung cancer from lung benign disease, normal control, and gastrointestinal cancers. ...
Comments