基于深度学习的非小细胞肺癌耐药性研究
Research on Drug Resistance of Non-Small Cell Lung Cancer Based on Deep Learning
DOI: 10.12677/CSA.2020.1010195, PDF,    科研立项经费支持
作者: 韦 涛*, 李璧江, 穆 亮, 梁 婵：广西大学计算机与电子信息学院，广西 南宁；张学军#：广西大学计算机与电子信息学院，广西 南宁;广西多媒体通信与网络技术重点实验室，广西 南宁
关键词: 深度学习；非小细胞肺癌(NSCLC)；耐药性预测；医学图像序列；迁移学习；Deep Learning； Non-Small Cell Lung Cancer (NSCLC)； Drug Resistance； Serial Medical Imaging； Transfer Learning
摘要: 非小细胞肺癌(NSCLC)是一种最常见的肺部癌症，临床上主要通过分子靶向药物进行治疗。随着时间的推移，NSCLC患者很容易对药物产生不同类型的耐药性，增加了临床治疗的难度。为了捕捉肿瘤不同时间点的生物学特性来预测患者的耐药性类型，本文提出了一种基于卷积神经网络(CNN)和循环神经网络(RNN)的深度学习模型，CNN用于提取不同时间点的肿瘤图像特征，并将这些特征输入到RNN中做进一步的纵向分析。168例NSCLC患者数据按3:1的比例划分为训练组和测试组，结合迁移学习完成模型构建。经检验，模型在测试集上的预测精度为79.16%，表明了将本方法用于预测NSCLC患者的耐药性类型具有一定的可行性。
Abstract: Non-small cell lung cancer (NSCLC) is one of the common lung cancers and mainly treated clinically through Molecular targeted drugs. However, NSCLC patients are prone to develop different types of drug resistance, which increase the difficulties of clinical treatment. This study proposed a deep learning model based on Convolutional Neural Network (CNN) and Recurrent Neural Network (RNN) to predict the type of drug resistance of NSCLC patients by capturing the biological characteristics of tumors at different time points. CNN is used to extract tumor image features at different time points, and output these features into RNN for further longitudinal analysis. A total of 168 NSCLC patients are split 3:1 into training cohort and test cohort, which are combined with transfer learning to build the model. Experimental results show that the model achieves 79.16% accuracy on test set, which indicates that this method has certain feasibility for predicting the resistance type of NSCLC patients.
文章引用：韦涛, 李璧江, 穆亮, 梁婵, 张学军. 基于深度学习的非小细胞肺癌耐药性研究[J]. 计算机科学与应用, 2020, 10(10): 1853-1862. https://doi.org/10.12677/CSA.2020.1010195

参考文献

[1] Chen, W., Zheng, R., Baade, P.D., et al. (2016) Cancer Statistics in China, 2015. CA: A Cancer Journal for Clinicians, 66, 115-132. [Google Scholar] [CrossRef] [PubMed]
[2] 石远凯, 孙燕, 丁翠敏, 等. 中国埃克替尼治疗非小细胞肺癌专家共识[J]. 中国肺癌杂志, 2015, 18(7): 397-400.
[3] Costa, D.B., Kobayashi, S., Tenen, D.G., et al. (2007) Pooled Analysis of the Prospective Trials of Gefitinib Monotherapy for EGFR-Mutant Non-Small Cell Lungcancers. Lung Cancer, 58, 95-103. [Google Scholar] [CrossRef] [PubMed]
[4] Miller, V.A., Riely, G.J., Zakowski, M.F., et al. (2008) Molec-ular Characteristics of Bronchioloalveolar Carcinoma and Adenocarcinoma, Bronchioloalveolar Carcinoma Subtype, Pre-dict Response to Erlotinib. Journal of Clinical Oncology, 26, 1472-1478. [Google Scholar] [CrossRef] 
[5] Jackman, D., Pao, W., Riely, G.J., et al. (2010) Clinical Definition of Acquired Resistance to Epidermal Growth Factor Receptor Tyrosine Kinase Inhibitors in Non-Small-Cell Lung Cancer. Journal of Clinical Oncology, 28, 357-360. [Google Scholar] [CrossRef] 
[6] Yang, J.J., Chen, H.J., Yan, H.H., et al. (2013) Clinical Modes of EGFR Tyrosine Kinase Inhibitor Failure and Subsequent Management in Advanced Non-Small Cell Lung Cancer. Lung Cancer, 79, 33-39. [Google Scholar] [CrossRef] [PubMed]
[7] LeCun, Y., Bengio, Y. and Hinto, G. (2015) Deep Learning. Nature, 521, 436-444. [Google Scholar] [CrossRef] [PubMed]
[8] Shin, H.-C., Roth, H.R., Gao, M., Lu, L., Xu, Z., Nogues, I., et al. (2016) Deep Convolutional Neural Networks for Computer-Aided Detection: CNN Architectures, Dataset Characteristics and Transfer Learning. IEEE Transactions on Medical Imaging, 35, 1285-1298. [Google Scholar] [CrossRef] 
[9] Dandil, E., Cakiroglu, M., Eksi, Z., Ozkan, M., Kurt, O.K. and Canan, A. (2014) Artificial Neural Network-Based Classification System for Lung Nodules on Computed Tomography Scans. 6th International Conference on Soft Computing and Pattern Recognition, Tunis, 11-14 August 2014, 382-386. [Google Scholar] [CrossRef] 
[10] Wang, S., Shi, J., Ye, Z., et al. (2019) Predicting EGFR Mutation Status in Lung Adenocarcinoma on CT Image Using Deep Learning. European Respiratory Journal, 53, Arti-cle ID: 1800986. [Google Scholar] [CrossRef] [PubMed]
[11] Ioffe, S. and Szegedy, C. (2015) Batch Normalization: Accel-erating Deep Network Training by Reducing Internal Covariate Shift. Proceedings of the 32nd International Conference on Machine Learning, Lille, Volume 37, 448-456.
[12] Srivastava, N., Hinton, G., Krizhevsky, A., et al. (2014) Drop-out: A Simple Way to Prevent Neural Networks from Overfitting. The Journal of Machine Learning Research, 15, 1929-1958.
[13] Sainath, T.N., Mohamed, A., Kingsbury, B., et al. (2013) Deep Convolutional Neural Networks for LVCSR. Proceeding of the IEEE International Conference on Acoustics, Speech and Signal Processing, Vancouver, 8614-8618. [Google Scholar] [CrossRef] 
[14] Szegedy, C., Vanhoucke, V., Ioffe, S., Shlens, J. and Wojna, Z. (2016) Rethinking the Inception Architecture for Computer Vision. 2016 IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, 2818-2826. [Google Scholar] [CrossRef] 
[15] Fawcett, T. (2006) An Introduction to ROC Analysis. Pattern Recognition Letters, 27, 861-874. [Google Scholar] [CrossRef]