|
|
Pulmonary Nodule Detection Algorithm Based on Faster-RCNN |
Song Shangling1, Yang Yang2*, Li Xia2, Feng Hao2 |
1(The Second Hospital of Shandong University, Jinan 250033, China)
2(School of Information Science and Engineering, Shandong University, Qingdao 266237, Shandong, China) |
|
|
Abstract Aiming to overcome the problems of individual differences and the homograph in the detection of pulmonary nodules, this paper presented a method of automatic recognition of pulmonary nodules based on Faster-RCNN. By comparing the adaptability of the current in-depth learning model, a general strategy was proposed to continuously improve the detection rate of pulmonary nodules with the increase of the number of samples. First of all, the hardware and software environment for deep learning was built; next, the data interface was set to match with the network interface of Faster-RCNN. Secondly, the single-category classification network of Faster-RCNN was set up and the parameters were adjusted. Thirdly, the pulmonary nodules database containing 2 000 patients was utilized to train different feature extraction models (including ZF and VGG models), and the features of CT pictures in different networks were calculated. The test results, missed detection rate and detection accuracy were evaluated. Finally, the influence of different training numbers and data augmentation types on the final detection accuracy was analyzed. The accuracy rate of ZF model was 90.82%, the variance of accuracy rate was 13.30%; the detection accuracy of VGG model was 87.02%, the variance of accuracy rate was 37.10%. Taking into account the balance between the missed detection rate and detection accuracy rate, the ZF model showed small fluctuation variance, a slight low accuracy, and high detection precision. Therefore, the ZF model for pulmonary nodules was better than VGG model. Our proposed lung nodule detection technology has a good theoretical value and engineering application value.
|
Received: 10 January 2019
|
|
Corresponding Authors:
*E-mail: yyang@sdu.edu.cn
|
|
|
|
[1] Vaughn B, Jonathan R, Alexander M. Architecture and CAD for deep-submicron FPGAs [M]//Science & Business Media. Des Moines: Springer Science & Business Media, 2012.
[2] Ginneken BV, Setio AAA, Jacobs C, et al. Off-the-shelf convolutional neural network features for pulmonary nodule detection in computed tomography scans [C]// International Symposium on Biomedical Imaging. Piscataway:IEEE, 2015: 286-289.
[3] Chang YC. RGB calibration for color image analysis in machine vision [J]. IEEE Trans Image Process, 1996, 5(10):1414-1422.
[4] Bergmann F, Segal R, Rimon S. A new type of esterase in hog-kidney extract [J]. Biochemical Journal, 1957, 67(3):481-486.
[5] Hadler NM. The sociopolitical climate surrounding low back pain (LBP) [J]. Journal of Occupational Medicine, 1979, 21(10):681-682.
[6] John J, Mini M G. Multilevel thresholding based segmentation and feature extraction for pulmonary nodule detection [J]. Procedia Technology, 2016, 24:957-963.
[7] Alam M, Sankaranarayanan G, Devarajan V. Lung nodule detection and segmentation using a patch-based multi-atlas method [C] //International Conference on Computational Science and Computational Intelligence. Piscataway: IEEE, 2017:23-28.
[8] Wang J, Cheng Y, Guo C, et al. A new pulmonary nodules detection scheme utilizing region grow and adaptive fuzzy C-means clustering [J]. Journal of Medical Imaging & Health Informatics, 2015, 5(8):1941-1946.
[9] Li X, Yang Y, Xiong H, et al. Pulmonary nodules detection algorithm based on robust cascade classifier for CT images [C]// The 29th Chinese Control and Decision Conference, Chongqing: IEEE, 2017:231-235.
[10] Qi D, Hao C, Yu L, et al. Multilevel contextual 3-D CNNs for false positive reduction in pulmonary nodule detection [J]. IEEE Transactions on Biomedical Engineering, 2017, 64(7):1558-1567.
[11] Lin TY, Dollár P, Ross G, et al. Feature pyramid networks for object detection [C]// IEEE Conference on Computer Vision and Pattern Recognition. Honolulu: IEEE, 2017:936-944.
[12] Ren Shaoqing, He Kaiming, Girshick R, et al. Faster R-CNN: Towards real-time object detection with region proposal networks [J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2017, 39(6):1137-1149.
[13] Patroumpas K. Multi-scale window specification over streaming trajectories [J]. Journal of Spatial Information Science, 2013, 7: 45-75.
[14] Burt PJ, Adelson EH. The Laplacian pyramid as a compact image code [J]. IEEE Transactions on Communications, 1983, 31(4): 532-540.
[15] Zeiler MD, Fergus R. Visualizing and understanding convolutional networks [C]//European Conference on Computer Vision. Zurich: Springer, 2014:818-833.
[16] Sainath TN, Kingsbury B, Saon G, et al. Deep convolutional neural networks for large-scale speech tasks [J]. Neural Networks, 2015, 64:39-48.
[17] Xun L, Wang Y, Cheng W, et al. A New equation to estimate glomerular filtration rate in Chinese elderly population [J]. PLoS ONE, 2013, 8(11): e79675.
[18] Frid-Adar M, Klang E, Amitai M, et al Synthetic data augmentation using GAN for improved liver lesion classification [C]// IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018), Washington, DC: IEEE, 2018:289-293. |
[1] |
Yu Hui, Wang Shuo, Li Xinrui, Deng Chenyang, Sun Jinglai, Zhang Lixin, Cao Yuzhen. Algorithm Study of Real-Time Detection of Sleep Apnea-Hypopnea Event Based on Long-Short Term Memory-Convolutional Neural Network[J]. Chinese Journal of Biomedical Engineering, 2020, 39(3): 303-310. |
[2] |
Yu Hui, Du Peipei, Liu Xiang, Liu Zheng, Zhu Xianfeng, Cao Yuzhen. Intelligent Classification and Identification of Composite Colonies Based on Convolutional Neural Network[J]. Chinese Journal of Biomedical Engineering, 2020, 39(1): 26-32. |
|
|
|
|