|
|
|
| Size-Adaptive Deep Neural Networks Based Pulmonary Nodule Detection in CT Scans |
| Ai Qi1, Wang Jun2, Ren Fuquan3*, Weng Wencai1, Yu Qiulei1 |
1(Dalian University Affiliated Xinhua Hospital, Dalian 116021, Liaoning, China)
2(Zhejiang University City College, Hangzhou 310015, China)
3(School of Science, Yanshan University, Qinhuangdao 066004, Hebei, China) |
|
|
|
|
Abstract Computed tomography (CT) screening for pulmonary nodules is an important method for early diagnosis of lung cancer. However, the automatic detection of pulmonary nodules, especially small nodules, is still challenging due to the large differences in shape, size and location of pulmonary nodules. To achieve highly sensitive detection of pulmonary nodules, in this paper, a new computer-aided detection system for pulmonary nodules detection was proposed. The system adopted two new strategies: size adaptive candidate test (SACD) and size adaptive false positive reduction (SAFPR). First, SACD combined deep and shallow convolution features to construct advanced features and detected CT images to obtain the location and size information of the region of interest. Then, the detection results were sent to three parallel sub-networks for screening different sizes of nodules, so as to refine the detection results of SACD and improved the accuracy and robustness of the computer-aided detection system. The results on the LIDC-IDRI dataset (1186 nodules) demonstrated that the proposed system achieved the high sensitivity of 96% at 1 FPs/scan, which was superior or comparable to the state-of-the-art systems, while in an independent dataset containing 430 nodules, the detection sensitivity of the system was 69.53% at 0.3 FPs/ scan for the nodules with the size of 3.89±2.34 mm, which was comparable to the human screening results of two experienced radiologists, indicating that the system has certain clinical application value.
|
|
Received: 09 December 2020
|
|
|
|
|
|
[1] Bray F, Ferlay J, Soerjomatarm I, et al. Global cancer statistics 2018: Globocan estimates of incidence and mortality worldwide for 36 cancers in 185 countries [J]. CA Cancer J Clin, 2018, 68(6): 394-424.
[2] Aberle DR, Adams A M,Berg CD, et al. Reduced lung-cancer mortality with low-dose computed tomography screening [J]. N Engl J Med, 2011, 365(5): 395-409.
[3] Liang Mingzhu, Tang Wei, Xu Mingdong, et al. Low-dose CT screening for lung cancer: Computer-aided detection of missed lung cancers [J]. Radiology, 2016, 281(1): 279-288.
[4] Henschke CI, Yankelevitz DF, Yip R, et al. Lung cancers diagnosed at annual ct screening: volume doubling times [J]. Radiology, 2012, 263(2): 578-583.
[5] Xu Mingdong, Yip R, Smith JP, et al. Retrospective review of lung cancers diagnosed in annual rounds of CT screening [J]. AJR Am J Roentgenol, 2014, 203(5): 965-972.
[6] Ye Xujiong, Lin Xinyu, Dehmeshki J, et al. Shape-based computer-aided detection of lung nodules in thoracic CT images [J]. IEEE Trans Biomed Eng, 2009, 56(7): 1810-1820.
[7] Murphy K, van Ginneken B, Schilham AM, et al. A large-scale evaluation of automatic pulmonary nodule detection in CT using local image features and k-nearest-neighbour classification [J]. Med Image Anal, 2009, 13(5): 757-770.
[8] Jacobs C, van Rikxoort EM, Twellmann T, et al. Automatic detection of subsolid pulmonary nodules in thoracic computed tomography images [J]. Med Image Anal, 2014, 18(2): 374-384.
[9] Setio AA, Jacobs C, Gelderblom J, et al. Automatic detection of large pulmonary solid nodules in thoracic CT images [J]. Med Phys, 2015, 42(10): 5642-5653.
[10] 巩萍, 程玉虎, 王雪松. 基于语义属性的肺结节良恶性分类 [J]. 电子学报, 2015, 43(12): 2476-2483.
[11] Ciompi F, Jacobs C, Scholten ET, et al. Bag-of-frequencies: A descriptor of pulmonary nodules in computed tomography images [J]. IEEE Trans Med Imag, 2015, 34(4): 962-973.
[12] Ronneberger O, Fischer P, and Brox T. U-net: convolutional networks for biomedical image segmentation [C]// Medical Image Computing and Computer Assisted Interventions Conference. Munich: Springer (LNCS), 2015, 9351: 234-241.
[13] Long J, Shelhamer E, and Darrell T. Fully convolutional networks for semantic segmentation [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015, 39(4): 640-651.
[14] Girshick R. Fast R-CNN [C]//IEEE International Conference on Computer Vision. Santiago: IEEE Press, 2015: 1440-1448.
[15] Ren Shaoqing, He Kaiming, Girshick R et al. Faster R-CNN: Torwards real-time object detection with region proposal networks [J]. IEEE TPAMI, 2017, 39(6): 1137-1149.
[16] He Kaiming, Gkioxari G, Dollar P, et al. Mask R-CNN [C]//IEEE International Conference on Computer Vision. Venice: IEEE Press, 2017: 2980-2988.
[17] Liu Wei, Anguelov D, Erhan D, et al. SSD: Single shot multibox detector [C]//Proceedings of European Conference on Computer Vision. Amsterdam: Spring (LNCS), 2016, 9905: 21-37.
[18] Lin Tsungyi, Dollar P, Girshick R, et al. Feature pyramid networks for object detection [C]//Proceedings of IEEE Conference Computer Vision and Pattern Recognition. Honolulu: IEEE Press, 2017: 936-944.
[19] Kong Tao, Yao Anbang, Chen Yurong, et al. Hypernet: towards accurate region proposal generation and joint object detection [C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas: IEEE Press, 2016: 845-853.
[20] He Kaiming, Zhang Xiangyu, Ren Shaoqing, et al. Deep residual learning for image recognition [C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas: IEEE Press, 2016: 770-778.
[21] Szegedy C, Sergey loffe, Vanhoucke V, et al. Inception-v4, inception-resnet and the impact of residual connections on learning [C]//Proceedings of AAAI Conference on Artificial Intelligence. San Francisco: AAAI Press, 2017: 4278-4284.
[22] Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition [EB/OL]. http://arxiv.org/abs/1409.1556, 2016-04-10/2020-07-16
[23] Zeiler MD, and Fergus R. Visualizing and understanding convolutional neural networks [C]//Proceedings of European Conference on Computer Vision. Zurich: IEEE Press, 2014: 818-833.
[24] Shao Ling, Zhu Fan, Li Xuelong. Transfer learning for visual categorizations: a survey [J]. IEEE Transactions on Neural Networks and Learning Systems, 2015, 26(5): 1019-1034.
[25] Dou Qi, Chen Hao, Lequan Yu, et al. Multilevel contextual 3-D CNNs for false positive reduction in pulmonary nodule detection [J]. IEEE Trans Biomed Eng, 2017 64(7): 1558-1567.
[26] 张佳嘉,张小洪. 多分支卷积神经网络肺结节分类方法及其可解释性 [J]. 计算机科学, 2020, 47(09): 135-140.
[27] Jin Hongsheng, Li Zongyao, Tong Ruofeng, et al. A deep 3D residual CNN for false positive reduction in pulmonary nodule detection [J]. Med Phys, 2018, 45(5): 2097-2107.
[28] Dou Qi, Chen Hao, Jin Yueming, et al. Automated pulmonary nodule detection via 3D convnets with online sample filtering and hybrid-loss residual learning [C]//Medical Image Computing and Computer Assisted Interventions Conference. Quebec: Springer (LNCS), 2017, vol 10435: 630-638.
[29] Chen Sihong, Qin Jing, Ji, Xing, et al. Automatic scoring of multiple semantic attributes with multi-task feature leverage: a study on pulmonary nodules in CT images [J]. IEEE Trans Med Imag, 2017, 36(3): 802-814.
[30] Setio AA, Traverso A, de Bel T, et al. Validation, comparison, and combination of algorithms for automatic detection of pulmonary nodules in computed tomography images: the luna16 challenge [J]. Med Image Anal, 2017, 42: 1-13.
[31] Zhu Wentao, Liu Chaochun, Fan Wei, et al. Deeplung: deep 3D dual path nets for automated pulmonary nodule detection and classification [C]//IEEE Winter Conference on Applications of Computer Vision. Lake Tahoe: IEEE Press, 2018: 673-681.
[32] 宋尚玲, 杨阳, 李夏, 等. 基于Faster-RCNN的肺结节检测算法 [J]. 中国生物医学工程学报, 2020, 39(2): 129-136.
[33] Alibaba cloud computing Co, Ltd. TIANCHI challenge 2017, first season: lung nodule detection [DB/OL]. https://tianchi.aliyun.com/competition/entrance/231601, 2017-09-26/2020-6-10.
[34] Jacobs C, Setio AA, Traverso A, et al. LUNA-16: lung nodule analysis 2017 [EB/OL]. https://luna16. grand - challenge.org, 2016-01-14/2020-06-10.
[35] Armato III AG, McLennan G, Bidaut L, et al. The lung image database consortium (LIDC) and image database resource initiative (IDRI): A completed reference database of lung nodules on CT scans [J]. Med Phys, 2011, 38(2): 915-931.
[36] Boureau YL, Ponce J, LeCun Y. A theoretical analysis of feature pooling in visual recognition [C]//Proceedings of Internation Conference on Machine Learning. Haifa: DBLP, 2010: 111-118.
[37] Lin Yun, Cheng Ming-Ming, Hu Xiaowei et al. Richer convolutional features for edge detection [J]. IEEE Trans Pattern Anal Mach Intell, 2019, 41(8): 1939-1946.
[38] Girshick R, Iandola F, Darrell T et al. Deformable part models are convolutional neural networks [C]//Proceedings of IEEE Conference on Comput Vision and Pattern Recognition. Boston: IEEE Press, 2015: 437-446.
[39] Wan Lin, Zeiler M, Zhang Sixin, et al. Regularization of neural networks using dropconnect [C]//Proceedings of International Conference on Machine Learning. Atlanta: International Machine Learning Society, 2013: 1058-1066.
[40] DeLuca PM, Wambersie A, Whitmore GF. Receiver operating characteristic analysis in medical imaging [J]. Journal of the ICRU, 2008, 8(1): 1-62. |
|
|
|
