基于卷积神经网络的子宫内膜癌分类问题
Classification of Endometrial Carcinoma Based on Convolutional Neural Network
DOI: 10.12677/CSA.2021.116180, PDF,    国家自然科学基金支持
作者: 钟 滢, 张 悦：东北大学理学院，辽宁 沈阳
关键词: 子宫内膜癌；基因表达；卷积神经网络；归一化；卷积核；Endometrial Cancer； Gene Expression； Convolutional Neural Network； Normalization； Convolution Kernel
摘要: 在本文中，讨论了基于卷积神经网络(CNN)对87位女性子宫内膜基因表达样本的分类问题。首先，删除掉缺失数据对应的基因，计算信噪比来过滤不相关的基因。然后，将每个指标相应的数据放入CNN中求出分类准确率。之后对每个指标进行归一化处理，同样通过CNN得到4个指标组合的分类准确率。最后，应用下三角矩阵和上三角零元素处理来改进初始化卷积核。后者有效地提高了训练集以及测试集的分类准确率。
Abstract: In this paper, the classification of 87 female endometrial gene expression samples based on convolutional neural network (CNN) is discussed. First, the genes corresponding to the missing data were deleted and the signal-to-noise ratio was calculated to filter out the unrelated genes. Then, the corresponding data of each indicator is put into CNN to calculate the classification accuracy rate. Then each indicator was normalized, and the classification accuracy rate of the four indicators combined was also obtained through CNN. Finally, the lower triangular matrix and the upper triangular zero element processing are used to improve the initial convolution kernel. The latter can effectively improve the classification accuracy rate of training set and test set.
文章引用：钟滢, 张悦. 基于卷积神经网络的子宫内膜癌分类问题[J]. 计算机科学与应用, 2021, 11(6): 1747-1754. https://doi.org/10.12677/CSA.2021.116180

参考文献

[1] 黄磊. 癌症分类中基因选择的收缩特征选择算法研究[D]: [硕士学位论文]. 长沙: 湖南大学, 2015.
[2] Golub, T.R., Slonim, D.K., Tamayo, P., et al. (1999) Molecular Classification of Cancer: Class Discovery and Class Prediction by Gene Expression Monitoring. Science, 286, 531-537. [Google Scholar] [CrossRef] [PubMed]
[3] Alon, U., Barkai, N., Notterman, D.A., et al. (1999) Broad Patterns of Gene Expression Revealed by Clustering Analysis of Tumor and Normal Colontissues Probed by Oligonucleotide Arrays. Proceedings of the National Academy of Sciences of the United States of America, 96, 6745-6750. [Google Scholar] [CrossRef] [PubMed]
[4] Armstrong, S.A., Staunton, J.E., Silverman, L.B., et al. (2002) MLL Translocations Specify a Distinct Gene Expression Profile That Dis-tinguishes a Unique Leukemia. Nature Genetics, 30, 41-47. [Google Scholar] [CrossRef] [PubMed]
[5] Hu, H.P., Niu, Z.J., Bai, Y.P. and Tan, X.H. (2015) Cancer Classification Based on Gene Expression Using Neural Networks. Genetics and Molecular Research, 14, 17605-17611. [Google Scholar] [CrossRef] 
[6] Peterson, C. and Ringner, M. (2003) Analyzing Tumor Gene Expression Profiles. Artificial Intelligence in Medicine, 28, 59-74. [Google Scholar] [CrossRef] 
[7] Zhang, J., Hu, X.G. and Zhang, Y.H. (2012) K-Split Lasso: An Effective Feature Selection Method for Tumor Gene Expression Data. Journal of Frontiers of Computer Science and Technology, 6, 1136-1143.
[8] Hu, H.P., Wang, H.Y., Bai, Y.P. and Liu, M.X. (2019) Determination of Endometrial Carcinoma with Gene Expression Based on Optimized Elman Neural Network. Applied Mathematics and Computation, 341, 204-214. [Google Scholar] [CrossRef] 
[9] Tirumala, S.S. and Narayanan, A. (2019) Classification and Diag-nostic Prediction of Prostate Cancer Using Gene Expression and Artificial Neural Networks. Neural Computing and Ap-plications, 31, 7539-7548. [Google Scholar] [CrossRef] 
[10] 胡红萍, 高帅, 孙强, 白艳萍. 基于基因表达子宫内膜癌的分类[J]. 数学的实践与认识, 2017, 47(18): 111-115.
[11] Gunavathi, C., Sivasubramanian, K., Keerthika, P. and Para-masivam, C. (2020) A Review on Convolutional Neural Network Based Deep Learning Methods in Gene Expression Data for Disease Diagnosis. Materials Today: Proceedings, 45, 2282-2285. [Google Scholar] [CrossRef] 
[12] Maharjan, S., Alsadoon, A., Prasad, P.W.C., et al. (2020) A Novel Enhanced Softmax Loss Function for Brain Tumour Detection Using Deep Learning. Journal of Neuroscience Methods, 330, 108520. [Google Scholar] [CrossRef] [PubMed]
[13] Lecun, Y., Boser, B., Denker, J.S., et al. (1989) Backpropa-gation Applied to Handwritten Zip Code Recognition. Neural Computation, 1, 541-551. [Google Scholar] [CrossRef] 
[14] Zou, X.W., Wang, Z.D., Li, Q. and Sheng, W.G. (2019) Integration of Residual Network and Convolutional Neural Network along with Various Activation Functions and Global Pooling for Time Series Classification. Neurocomputing, 367, 39-45. [Google Scholar] [CrossRef] 
[15] Sezer, O.B. and Ozbayoglu, A.M. (2018) Algorithmic Financial Trading with Deep Convolutional Neural Networks: Time Series to Image Conversion Approach. Applied Soft Compu-ting, 70, 525-538. [Google Scholar] [CrossRef] 
[16] 刘萌萌. 基于深度学习的微小元件姿态识别算法研究[D]: [硕士学位论文]. 郑州: 郑州大学, 2020.
[17] 吴新建. 基于卷积神经网络的图像标注算法研究[D]: [硕士学位论文]. 苏州: 苏州大学, 2019.
[18] Liu, J.L., Chao, F., Lin, C.-M., et al. (2021) DK-CNNs: Dynamic Kernel Convolutional Neural Networks. Neurocomputing, 422, 95-108. [Google Scholar] [CrossRef] 
[19] Li, G.Q., Shen, X.Z., Li, J.J. and Wang, J.Y. (2021) Diagonal-Kernel Convolutional Neural Networks for Image Classification. Digital Signal Processing, 108, 102898. [Google Scholar] [CrossRef] 
[20] 陈文祺. 基于深度学习的手机移动端视线跟踪算法研究[D]: [硕士学位论文]. 哈尔滨: 哈尔滨工业大学, 2019.
[21] Gu, J.X., Wang, Z.H., Jason, K., et al. (2018) Recent Advances in Convolution Neural Networks. Pattern Recognition, 77, 354-377. [Google Scholar] [CrossRef] 
[22] Jha, S., Son, L.H., Kumar, R., et al. (2019) Neutrosophic Image Segmentation with Dice Coefficients. Measurement, 134, 762-772. [Google Scholar] [CrossRef] 
[23] Finch, B.K. and Beck, A.N. (2011) Socio-Economic Sta-tus and z-Score Standardized Height-for-Age of U.S.-Born Children (Ages 2 - 6). Economics and Human Biology, 9, 272-276. [Google Scholar] [CrossRef] [PubMed]