|
|
基于3D CMFF深度网络模型的肺结节分类
|
Abstract:
当今基于深度学习的肺小结节分类成为了研究热点。然而，现有的网络模型大都精度过低，且只能处理2D肺结节切片，无法在肺部CT中学习到3D特征。为了应对这些挑战，提出了一个具有高精度的多尺度特征融合网络模型3DCMFF。3DCMFF的主干基于3DECABlock，可以在提高网络的特征表达能力。模型中的3D PSA模块，能够提取特征信息更丰富的3D精细化特征图。多尺度特征融合将阶段2~4层输出的特征图进行融合，在几乎不增加计算量的同时提升精度。在LUNA16数据集上进行综合消融实验和对比实验中，3DCMFF取得了93.58的精度。实验结果表明，我们的方法对分类精度有着可观的提升，并且和现有的先进模型相比较，3DCMFF具有精度高，鲁棒性优秀的特点，并能够给医生提供辅助诊断。
Nowadays, deep learning based classification of pulmonary nodules has become a research hotspot. However, most existing network models have low accuracy, as they can only handle 2D lung nodule slices and cannot learn 3D features in lung CT. To address these challenges, a high-precision mul-ti-scale feature fusion network model 3D CMFF has been proposed. The backbone of 3D CMFF is based on 3D ECA Block, which can improve the feature expression ability of the network. The 3D PSA module in the model can extract 3D refined feature maps with richer feature information. Multi scale feature fusion integrates feature maps output from stages 2~4, improving accuracy with al-most no increase in computational complexity. In the comprehensive ablation and comparative ex-periments conducted on the LUNA16 dataset, 3D CMFF achieved an accuracy of 93.58. The experi-mental results show that our method has significantly improved classification accuracy, and com-pared with existing advanced models, 3D CMFF has the characteristics of high accuracy and excel-lent robustness, which can provide auxiliary diagnosis for doctors.
| [1] | Siegel, R.L., Miller, K.D., Fuchs, H.E., et al. (2021) Cancer Statistics, 2021 (vol 71, pg 7, 2021). CA: A Cancer Journal for Clinicians, 71, 359-359. |
| [2] | Armato, S.G., Mclennan, G., Bidaut, L., et al. (2011) The Lung Image Database Con-sortium (LIDC) and Image Database Resource Initiative (IDRI): A Completed Reference Database of Lung Nodules on CT Scans. Medical Physics, 38, 915-931. https://doi.org/10.1118/1.3528204 |
| [3] | Huang, P.W., Lin, P.L., Lee, C.H., et al. (2013) A Classification System of Lung Nodules in CT Images Based on Fractional Brownian Motion Model. IEEE International Conference on System Science and Engineering (ICSSE), Budapest, 4-6 July 2013, 37-40. https://doi.org/10.1109/ICSSE.2013.6614710 |
| [4] | El-Baz, A., Nitzken, M., Khalifa, F., et al. (2011) 3D Shape Analysis for Early Diagnosis of Malignant Lung Nodules. 22nd International Conference on Information Processing in Medical Imaging (IPMI), Kloster Irsee, 3-8 July 2011, 772-783. https://doi.org/10.1007/978-3-642-22092-0_63 |
| [5] | Mao, Q., Zhao, S.G., Ren, L.J., et al. (2021) Intelligent Im-mune Clonal Optimization Algorithm for Pulmonary Nodule Classification. Mathematical Biosciences and Engineering, 18, 4146-4161. https://doi.org/10.3934/mbe.2021208 |
| [6] | Chen, H., Li, W. and Zhu, Y.Y. (2021) Improved Win-dow Adaptive Gray Level Co-Occurrence Matrix for Extraction and Analysis of Texture Characteristics of Pulmonary Nodules. Computer Methods and Programs in Biomedicine, 208, Article ID: 106263. https://doi.org/10.1016/j.cmpb.2021.106263 |
| [7] | Momoki, Y., Ichinose, A., Shigeto, Y., et al. (2022) Characteri-zation of Pulmonary Nodules in Computed Tomography Images Based on Pseudo-Labeling Using Radiology Reports. IEEE Transactions on Circuits and Systems for Video Technology, 32, 2582-2591. https://doi.org/10.1109/TCSVT.2021.3073021 |
| [8] | Bharti, M., Choudhary, J. and Singh, D.P. (2022) Detection and Classification of Pulmonary Lung Nodules in CT Images Using 3D Convolutional Neural Networks. 8th Interna-tional Conference on Advanced Computing and Communication Systems, ICACCS 2022, Coimbatore, 25-26 March 2022, 1319-1324.
https://doi.org/10.1109/ICACCS54159.2022.9785134 |
| [9] | Naik, A., Edla, D.R. and Kuppili, V. (2021) Lung Nodule Classification on Computed Tomography Images Using Fractalnet. Wireless Personal Communications, 119, 1209-1229. https://doi.org/10.1007/s11277-021-08258-w |
| [10] | Astaraki, M., Zakko, Y., Dasu, I.T., et al. (2021) Benign-Malignant Pulmonary Nodule Classification in Low-Dose CT with Convolutional Features. Physica Medi-ca-European Journal of Medical Physics, 83, 146-153.
https://doi.org/10.1016/j.ejmp.2021.03.013 |
| [11] | Shaffie, A., Soliman, A., Khalifeh, H.A., et al. (2020) A Compre-hensive Framework for Accurate Classification of Pulmonary Nodules. 2020 IEEE International Conference on Image Processing, ICIP 2020, Abu Dhabi, 25-28 September 2020, 408-412. https://doi.org/10.1109/ICIP40778.2020.9190909 |
| [12] | Sakshiwala and Singh, M.P. (2022) A New Framework for Multi-Scale CNN-Based Malignancy Classification of Pulmonary Lung Nodules. Journal of Ambient Intelligence and Humanized Computing, 14, 4675-4683.
https://doi.org/10.1007/s12652-022-04368-w |
| [13] | Muzammil, M., Ali, I., Haq, I.U., et al. (2021) Pulmonary Nod-ule Classification Using Feature and Ensemble Learning-Based Fusion Techniques. IEEE Access, 9, 113415-113427. https://doi.org/10.1109/ACCESS.2021.3102707 |
| [14] | Apostolopoulos, I.D., Apostolopoulos, D.J. and Panayiotakis, G.S. (2022) Solitary Pulmonary Nodule Malignancy Classification Utilising 3D Features and Semi-Supervised Deep Learning. 13th International Conference on Information, Intelligence, Systems and Applications, IISA 2022, Corfu, 18-20 July 2022, 1-6.
https://doi.org/10.1109/IISA56318.2022.9904334 |
| [15] | Fu, X.H., Bi, L., Kumar, A., et al. (2022) An Atten-tion-Enhanced Cross-Task Network to Analyze Lung Nodule Attributes in CT Images. Pattern Recognition, 126, Article ID: 108576. https://doi.org/10.1016/j.patcog.2022.108576 |
| [16] | Moreno, A., Rueda, A. and Martinez, F. (2022) A Multi-Scale Self-Attention Network to Discriminate Pulmonary Nodules. 19th IEEE International Symposium on Bio-medical Imaging, ISBI 2022, Kolkata, 28-31 March 2022, 1-4.
https://doi.org/10.1109/ISBI52829.2022.9761574 |
| [17] | Roy, R., Mazumdar, S. and Chowdhury, A.S. (2022) ADGAN: Attribute-Driven Generative Adversarial Network for Synthesis and Multiclass Classification of Pulmonary Nodules. IEEE Transactions on Neural Networks and Learning Systems. |
| [18] | Huang, H., Li, Y., Wu, R.Y., et al. (2022) Benign-Malignant Classification of Pulmonary Nodule with Deep Feature Optimization Framework. Biomedical Signal Processing and Control, 76, Article ID: 103701.
https://doi.org/10.1016/j.bspc.2022.103701 |
| [19] | He, W.L., Li, B., Liao, R.Q., et al. (2022) An ISHAP-Based In-terpretation-Model-Guided Classification Method for Malignant Pulmonary Nodule. Knowledge-Based Systems, 237, Article ID: 107778.
https://doi.org/10.1016/j.knosys.2021.107778 |
| [20] | Bonavita, I., Rafael-Palou, X., Ceresa, M., et al. (2020) Integra-tion of Convolutional Neural Networks for Pulmonary Nodule Malignancy Assessment in a Lung Cancer Classification Pipeline. Computer Methods and Programs in Biomedicine, 185, Article ID: 105172. https://doi.org/10.1016/j.cmpb.2019.105172 |
| [21] | Jiang, H.L., Shen, F.H., Gao, F., et al. (2021) Learning Efficient, Explainable and Discriminative Representations for Pulmonary Nodules Classification. Pattern Recognition, 113, Article ID: 107825.
https://doi.org/10.1016/j.patcog.2021.107825 |
| [22] | He, K.M., Zhang, X.Y., Ren, S.Q., et al. (2016) Deep Residual Learning for Image Recognition. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Ve-gas, 27-30 June 2016, 770-778.
https://doi.org/10.1109/CVPR.2016.90 |
| [23] | Wang, Q.L., et al. (2019) ECA-Net: Efficient Channel Attention for Deep Convolutional Neural Networks. |
| [24] | Zhu, W.T., Liu, C.C., Fan, W., et al. (2018) DeepLung: Deep 3D Dual Path Nets for Automated Pulmonary Nodule Detection and Classification. 18th IEEE Winter Conference on Applications of Computer Vision (WACV), Lake Tahoe, 12-15 March 2018, 673-681. https://doi.org/10.1109/WACV.2018.00079 |
