Journal of Computer Applications ›› 2022, Vol. 42 ›› Issue (5): 1424-1430.DOI: 10.11772/j.issn.1001-9081.2021050813
• Artificial intelligence • Previous Articles Next Articles
Xinlin XIE1,2, Yi XIAO3, Xinying XU3()
Received:
2021-05-17
Revised:
2021-09-26
Accepted:
2021-11-26
Online:
2022-03-08
Published:
2022-05-10
Contact:
Xinying XU
About author:
XIE Xinlin, born in 1990,Ph. D.,lecturer. His research interests include medical image processing,rough set.Supported by:
通讯作者:
续欣莹
作者简介:
谢新林(1990—),男,山西运城人,讲师,博士,CCF会员,主要研究方向:医学图像处理、粗糙集基金资助:
CLC Number:
Xinlin XIE, Yi XIAO, Xinying XU. Lung nodule classification algorithm based on neural network architecture search[J]. Journal of Computer Applications, 2022, 42(5): 1424-1430.
谢新林, 肖毅, 续欣莹. 基于神经网络架构搜索的肺结节分类算法[J]. 《计算机应用》唯一官方网站, 2022, 42(5): 1424-1430.
Add to citation manager EndNote|Ris|BibTeX
URL: http://www.joca.cn/EN/10.11772/j.issn.1001-9081.2021050813
算法 | Acc. | Sens. | Spec. | Pre. | AUC | F1 |
---|---|---|---|---|---|---|
NAS | 87.57 | 86.30 | 94.64 | 87.31 | 92.03 | 87.10 |
NAS+MS-SCAM | 88.04 | 85.08 | 95.73 | 88.24 | 92.59 | 88.51 |
NAS+integrated | 89.20 | 88.18 | 94.36 | 89.19 | 92.82 | 89.11 |
NAS+MS-SCAM+integrated | 90.86 | 87.78 | 95.37 | 88.74 | 93.42 | 89.19 |
Tab. 1 Ablation experimental results under different strategies
算法 | Acc. | Sens. | Spec. | Pre. | AUC | F1 |
---|---|---|---|---|---|---|
NAS | 87.57 | 86.30 | 94.64 | 87.31 | 92.03 | 87.10 |
NAS+MS-SCAM | 88.04 | 85.08 | 95.73 | 88.24 | 92.59 | 88.51 |
NAS+integrated | 89.20 | 88.18 | 94.36 | 89.19 | 92.82 | 89.11 |
NAS+MS-SCAM+integrated | 90.86 | 87.78 | 95.37 | 88.74 | 93.42 | 89.19 |
算法 | Acc. | Sens. | Spec. | Pre. | AUC | F1 |
---|---|---|---|---|---|---|
删除MS-SCAM模块 | 87.57 | 86.30 | 94.64 | 87.31 | 92.03 | 87.10 |
阶段3 | 87.78 | 86.43 | 94.98 | 87.60 | 92.26 | 87.51 |
阶段3+阶段4 | 87.89 | 85.51 | 95.42 | 87.97 | 92.46 | 87.98 |
阶段3+阶段4+阶段5 | 88.04 | 85.08 | 95.73 | 88.24 | 92.59 | 88.51 |
Tab. 2 Influence of number of MS-SCAM modules on classification performance
算法 | Acc. | Sens. | Spec. | Pre. | AUC | F1 |
---|---|---|---|---|---|---|
删除MS-SCAM模块 | 87.57 | 86.30 | 94.64 | 87.31 | 92.03 | 87.10 |
阶段3 | 87.78 | 86.43 | 94.98 | 87.60 | 92.26 | 87.51 |
阶段3+阶段4 | 87.89 | 85.51 | 95.42 | 87.97 | 92.46 | 87.98 |
阶段3+阶段4+阶段5 | 88.04 | 85.08 | 95.73 | 88.24 | 92.59 | 88.51 |
算法 | Acc. | Sens. | Spec. | Pre. | AUC | F1 |
---|---|---|---|---|---|---|
通道注意力+ 空间注意力 | 88.04 | 85.08 | 95.73 | 88.24 | 92.59 | 88.51 |
空间注意力+ 通道注意力 | 87.65 | 82.86 | 96.43 | 87.11 | 89.74 | 86.37 |
Tab. 3 Influence of order of channel attention and spatial attention on classification performance
算法 | Acc. | Sens. | Spec. | Pre. | AUC | F1 |
---|---|---|---|---|---|---|
通道注意力+ 空间注意力 | 88.04 | 85.08 | 95.73 | 88.24 | 92.59 | 88.51 |
空间注意力+ 通道注意力 | 87.65 | 82.86 | 96.43 | 87.11 | 89.74 | 86.37 |
算法 | Acc. | Sens. | Spec. | Pre. | AUC | F1 |
---|---|---|---|---|---|---|
NASpure | 87.04 | 83.48 | 90.63 | 85.44 | 89.01 | 84.59 |
NASpure+ 残差卷积cell | 87.55 | 84.47 | 92.13 | 86.25 | 90.77 | 86.56 |
NASpure+POP | 87.28 | 84.03 | 91.50 | 85.79 | 89.41 | 85.74 |
NASpure+残差 卷积cell+POP | 87.57 | 86.30 | 94.64 | 87.31 | 92.03 | 87.10 |
Tab. 4 Influence of neural network architecture search setting on classification performance
算法 | Acc. | Sens. | Spec. | Pre. | AUC | F1 |
---|---|---|---|---|---|---|
NASpure | 87.04 | 83.48 | 90.63 | 85.44 | 89.01 | 84.59 |
NASpure+ 残差卷积cell | 87.55 | 84.47 | 92.13 | 86.25 | 90.77 | 86.56 |
NASpure+POP | 87.28 | 84.03 | 91.50 | 85.79 | 89.41 | 85.74 |
NASpure+残差 卷积cell+POP | 87.57 | 86.30 | 94.64 | 87.31 | 92.03 | 87.10 |
多模型融合策略 | Acc./% | Sens./% | Spec./% | Pre./% | AUC/% | F1/% | 参数量/106 |
---|---|---|---|---|---|---|---|
加权投票 | 88.32 | 88.02 | 94.73 | 88.59 | 92.65 | 87.76 | 25.72 |
堆叠法 | 89.20 | 88.18 | 94.36 | 89.19 | 92.82 | 89.11 | 36.53 |
Tab. 5 Influence of multi-model fusion strategy on classification performance
多模型融合策略 | Acc./% | Sens./% | Spec./% | Pre./% | AUC/% | F1/% | 参数量/106 |
---|---|---|---|---|---|---|---|
加权投票 | 88.32 | 88.02 | 94.73 | 88.59 | 92.65 | 87.76 | 25.72 |
堆叠法 | 89.20 | 88.18 | 94.36 | 89.19 | 92.82 | 89.11 | 36.53 |
算法 | Acc./% | Sens./% | Spec./% | Pre./% | AUC/% | F1/% | 参数量/106 |
---|---|---|---|---|---|---|---|
Autoencoder | 80.29 | 73.00 | 85.00 | — | 86.00 | — | — |
MC-CNN | 87.14 | 77.00 | 93.00 | — | 93.00 | — | — |
Local-Global | 88.46 | 88.66 | — | — | 95.62 | — | — |
DeepLung | 90.44 | 81.42 | — | — | — | — | 141.57 |
NAS-lung | 90.77 | 85.37 | 95.04 | — | — | 89.29 | 16.84 |
MV-KBC | 91.60 | 86.52 | 94.00 | 87.75 | 95.70 | 87.13 | — |
本文算法 | 90.86 | 87.78 | 95.37 | 93.42 | 88.74 | 89.19 | 30.64 |
Tab. 6 Performance comparison of different classification algorithms on LIDC-IDRI dataset
算法 | Acc./% | Sens./% | Spec./% | Pre./% | AUC/% | F1/% | 参数量/106 |
---|---|---|---|---|---|---|---|
Autoencoder | 80.29 | 73.00 | 85.00 | — | 86.00 | — | — |
MC-CNN | 87.14 | 77.00 | 93.00 | — | 93.00 | — | — |
Local-Global | 88.46 | 88.66 | — | — | 95.62 | — | — |
DeepLung | 90.44 | 81.42 | — | — | — | — | 141.57 |
NAS-lung | 90.77 | 85.37 | 95.04 | — | — | 89.29 | 16.84 |
MV-KBC | 91.60 | 86.52 | 94.00 | 87.75 | 95.70 | 87.13 | — |
本文算法 | 90.86 | 87.78 | 95.37 | 93.42 | 88.74 | 89.19 | 30.64 |
1 | BRAY F, FERLAY J, SOERJOMATARAM I, et al. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries [J]. CA: A Cancer Journal for Clinicians, 2018, 68(6): 394-424. 10.3322/caac.21492 |
2 | 中华医学会呼吸病学分会肺癌学组,中国肺癌防治联盟专家组.肺结节诊治中国专家共识(2018年版)[J].中华结核和呼吸杂志,2018,41(10):763-771. 10.3760/cma.j.issn.1001-0939.2018.10.004 |
Lung Cancer Group of Chinese Thoracic Society of Chinese Medical Association, Expert Group of China Lung Cancer Prevention and Treatment Alliance. Interpretation of Chinese expert consensus on the diagnosis and treatment of pulmonary nodules (2018 version) [J]. Chinese Journal of Tuberculosis and Respiratory Diseases, 2018, 41(10): 763-771. 10.3760/cma.j.issn.1001-0939.2018.10.004 | |
3 | 李祥霞,李彬,田联房,等.基于放射影像组学和随机森林算法的肺结节良恶性分类[J].华南理工大学学报(自然科学版),2018,46(8):72-80. 10.3969/j.issn.1000-565X.2018.08.011 |
LI X X, LI B, TIAN L F, et al. Classification of benign and malignant pulmonary nodules based on radiomics and random forests algorithm [J]. Journal of South China University of Technology (Natural Science Edition), 2018, 46(8): 72-80. 10.3969/j.issn.1000-565X.2018.08.011 | |
4 | 强彦,裴博,赵涓涓,等.模糊支持向量机在肺结节良恶性分类中的应用[J].清华大学学报(自然科学版),2014,54(3):354-359. |
QIANG Y, PEI B, ZHAO J J, et al. Classification on pulmonary nodules based on a fuzzy support vector machine [J]. Journal of Tsinghua University (Science and Technology), 2014, 54(3): 354-359. | |
5 | 高峰,代美玲,祁瑾.基于Bootstrap-异质SVM集成学习的肺结节分类方法[J].天津大学学报(自然科学与工程技术版),2017,50(3):321-327. 10.11784/tdxbz201603019 |
GAO F, DAI M L, QI J. Classification of lung nodules by ensemble learning based on Bootstrap-heterogeneous SVM [J]. Journal of Tianjin University (Science and Technology), 2017, 50(3): 321-327. 10.11784/tdxbz201603019 | |
6 | 张婧,李彬,田联房,等.结合规则和SVM方法的肺结节识别[J].华南理工大学学报(自然科学版),2011,39(2):125-129,147. 10.3969/j.issn.1000-565X.2011.02.021 |
ZHANG J, LI B, TIAN L F, et al. Lung nodule recognition combining rule-based method and SVM [J]. Journal of South China University of Technology (Natural Science Edition), 2011, 39(2): 125-129, 147. 10.3969/j.issn.1000-565X.2011.02.021 | |
7 | SHEN W, ZHOU M, YANG F, et al. Multi-crop convolutional neural networks for lung nodule malignancy suspiciousness classification [J]. Pattern Recognition, 2017, 61: 663-673. 10.1016/j.patcog.2016.05.029 |
8 | XIE Y T, ZHANG J P, XIA Y, et al. Fusing texture, shape and deep model-learned information at decision level for automated classification of lung nodules on chest CT [J]. Information Fusion, 2018, 42: 102-110. 10.1016/j.inffus.2017.10.005 |
9 | XIE Y T, XIA Y, ZHANG J P, et al. Knowledge-based collaborative deep learning for benign-malignant lung nodule classification on chest CT [J]. IEEE Transactions on Medical Imaging, 2019, 38(4): 991-1004. 10.1109/tmi.2018.2876510 |
10 | ELSKEN T, METZEN J H, HUTTER F. Neural architecture search: a survey [J]. Journal of Machine Learning Research, 2019, 20: 1-21. 10.1007/978-3-030-05318-5_3 |
11 | LI X, ZHOU Y M, PAN Z, et al. Partial order pruning: for best speed/accuracy trade-off in neural architecture search [C]// Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2019: 9137-9145. 10.1109/cvpr.2019.00936 |
12 | 慧影医疗科技(北京)有限公司.一种基于多模型融合的肺结节检测方法及系统:中国,201910399988.6[P].2019-08-30. 10.1055/s-009-42979 |
Huiying Medical Technology (Beijing) Company Limited. A multi-model fusion-based method and system for detecting lung nodules: China, 201910399988.6 [P]. 2019-08-30. 10.1055/s-009-42979 | |
13 | LI F, CHENG D N, LIU M H. Alzheimer’s disease classification based on combination of multi-model convolutional networks [C]// Proceedings of the 2017 IEEE International Conference on Imaging Systems and Techniques. Piscataway: IEEE, 2017: 1-5. 10.1109/ist.2017.8261566 |
14 | JIANG H L, SHEN F H, GAO F, et al. Learning efficient, explainable and discriminative representations for pulmonary nodules classification [J]. Pattern Recognition, 2021, 113: Article No.107825. 10.1016/j.patcog.2021.107825 |
15 | KUMAR D, WONG A, CLAUSI D A. Lung nodule classification using deep features in CT images [C]// Proceedings of the 2015 12th Conference on Computer and Robot Vision. Piscataway: IEEE, 2015: 133-138. 10.1109/crv.2015.25 |
16 | AL-SHABI M, LAN B L, CHAN W Y, et al. Lung nodule classification using deep local-global networks [J]. International Journal of Computer Assisted Radiology and Surgery, 2019, 14(10): 1815-1819. 10.1007/s11548-019-01981-7 |
17 | ZHU W T, LIU C C, FAN W, et al. DeepLung: deep 3D dual path nets for automated pulmonary nodule detection and classification [C]// Proceedings of the 2018 IEEE Winter Conference on Applications of Computer Vision. Piscataway: IEEE, 2018: 673-681. 10.1109/wacv.2018.00079 |
[1] | Yongshuai LU, Yingjie TANG, Xinran MA. Low contrast filament sizing defect detection method of non-woven fabric based on deep feature fusion [J]. Journal of Computer Applications, 2022, 42(5): 1440-1446. |
[2] | Wei REN, Hexiang BAI. Multi-label image classification method based on global and local label relationship [J]. Journal of Computer Applications, 2022, 42(5): 1383-1390. |
[3] | Zhen QU, Kunting LI, Zhixi FENG. Remote sensing image scene classification based on effective channel attention [J]. Journal of Computer Applications, 2022, 42(5): 1431-1439. |
[4] | Yongru QIU, Guangle YAO, Jie FENG, Haoyu CUI. Single image de-raining algorithm based on semi-supervised learning [J]. Journal of Computer Applications, 2022, 42(5): 1577-1582. |
[5] | Haojie CHEN, Jiangting FAN, Yong LIU. Solving dynamic traveling salesman problem by deep reinforcement learning [J]. Journal of Computer Applications, 2022, 42(4): 1194-1200. |
[6] | Zumin WANG, Zhihao ZHANG, Jing QIN, Changqing JI. Review of mechanical fault diagnosis technology based on convolutional neural network [J]. Journal of Computer Applications, 2022, 42(4): 1036-1043. |
[7] | Changqing JI, Zhiyong GAO, Jing QIN, Zumin WANG. Review of image classification algorithms based on convolutional neural network [J]. Journal of Computer Applications, 2022, 42(4): 1044-1049. |
[8] | Yingjie WANG, Jiuqi ZHU, Zumin WANG, Fengbo BAI, Jian GONG. Review of applications of natural language processing in text sentiment analysis [J]. Journal of Computer Applications, 2022, 42(4): 1011-1020. |
[9] | Zhihua LIU, Wenjie CHEN, Aibin CHEN. Homologous spectrogram feature fusion with self-attention mechanism for bird sound classification [J]. Journal of Computer Applications, 2022, 42(4): 1260-1268. |
[10] | Jin ZHANG, Peiqi QU, Cheng SUN, Meng LUO. Safety helmet wearing detection algorithm based on improved YOLOv5 [J]. Journal of Computer Applications, 2022, 42(4): 1292-1300. |
[11] | Yongfeng DONG, Yahan DENG, Yao DONG, Yacong WANG. Survey of clustering based on deep learning [J]. Journal of Computer Applications, 2022, 42(4): 1021-1028. |
[12] | Tingxiu CHEN, Jianqin YIN. Audio visual joint action recognition based on key frame selection network [J]. Journal of Computer Applications, 2022, 42(3): 731-735. |
[13] | Qiujie SUN, Jinggui LIANG, Si LI. Chinese grammatical error correction model based on bidirectional and auto-regressive transformers noiser [J]. Journal of Computer Applications, 2022, 42(3): 860-866. |
[14] | Dejian WEI, Wenming WANG, Quanyu WANG, Haopan REN, Yanyan GAO, Zhi WANG. Improved 3D hand pose estimation network based on anchor [J]. Journal of Computer Applications, 2022, 42(3): 953-959. |
[15] | Wanying YU, Meiyu LIANG, Xiaoxiao WANG, Zheng CHEN, Xiaowen CAO. Student expression recognition and intelligent teaching evaluation in classroom teaching videos based on deep attention network [J]. Journal of Computer Applications, 2022, 42(3): 743-749. |
Viewed | ||||||
Full text |
|
|||||
Abstract |
|
|||||