# 基于卷积神经网络的细胞核智能分割研究A Nuclei Segmentation Research Based on Convolutional Neural Network

DOI: 10.12677/CSA.2018.811180, PDF, HTML, XML, 下载: 543  浏览: 1,528  国家自然科学基金支持

Abstract: In the pathological diagnosis of many diseases, the change of the shape and characteristics of the nucleus is an important symptom for the occurrence of the disease. Applying computer intelligence to segment the nuclei in the pathological tissue section can provide more advices for disease diagnosis. In this study, convolutional neural network was applied to the nuclei segmentation of breast cancer histopathological section image. After optical preprocessing the images, each of them was divided into multiple small images and used to train the improved AlexNet model. Then, the trained model is used in the nucleus segmentation of the test set. We divided the whole image into multiple small images, the small images were processed parallelly by the trained model, and finally integrated all the output to a whole nucleus segmentation image. The results show that the nucleus recognition rate in the training set reach to 92%. The trained model can accurately recognize all nuclei which are not labeled in the artificially labeled image, indicating that the model has success-fully learned the main features of the nucleus. Finally, the result of image segmentation in test set shows that the trained model successfully segmented the nucleus of pathological tissue slice image accurately and quickly, which proves that our method of cutting image to parallelly process and then integrating all outputs ensures both accuracy and calculation efficiency.

1. 引言

2. 方法

2.1. AlexNet深度学习网络

$ReLU\left(x\right)=\left\{\begin{array}{l}x,\text{\hspace{0.17em}}\text{\hspace{0.17em}}\text{\hspace{0.17em}}\text{ }x>0\\ 0,\text{\hspace{0.17em}}\text{\hspace{0.17em}}\text{\hspace{0.17em}}\text{ }\text{ }x\le 0\end{array}$

Figure 1. Improved Alexnet Neural Network structure Diagram

2.2. 数据集

2.3. 图片预处理

2.4. 细胞核分割

3. 结果

3.1. 图片预处理结果

Figure 2. The results of pretreatment. The figures on the left are the original tissue sections, and the figures on the right are tissue sections after optically transforming

3.2. 模型的训练

Figure 3. Recognition rate sampling diagram of neural network in 200,000 steps training process

Figure 4. The variation of the loss function per batch in 200,000 trials of training

3.3. 用训练后的模型进行细胞核分割

Figure 5. The left chart is the artificially scientific tag map for the tissue section, the middle chart is the original tissue section, and the right chart is the corresponding output of the neural network

Figure 6. The test set segmentation results of the trained neural network, the left charts are the segmentation results, and the right charts are the original tissue sections

4. 结论

NOTES

*通讯作者。

 [1] 杨槐. 乳腺癌病理学诊断研究[J]. 局解手术学, 2011, 20(2): 177-179. [2] 邓杨, 包骥. 数字病理中计算机辅助诊断研究展望[J]. 实用医院临床杂志, 2017, 14(5): 10-12. [3] 蔡海洋. 胃腺癌病理切片CAD系统的研究与实现[D]: [硕士学位论文]. 西安: 西安电子科技大学, 2008. [4] 项磊, 徐军. 基于HOG特征和滑动窗口的乳腺病理图像细胞检测[J]. 山东大学学报(工学版), 2015, 45(1): 37-44. [5] 张敏淑. 白细胞图像的特征提取与分类算法研究[D]: [硕士学位论文]. 杭州: 中国计量大学, 2016. [6] Xu, J., Xiang, L., Liu, Q., Gilmore, H., Wu, J., Tang, J. and Madabhushi, A. (2016) Stacked Sparse Autoencoder (SSAE) for Nuclei Detection on Breast Cancer Histopathology Images. IEEE Transactions on Medical Imaging, 35, 119-130. https://doi.org/10.1109/TMI.2015.2458702 [7] Win, K.Y., Choomchuay, S., Hamamoto, K. and Raveesunthornkiat, M. (2018) Artificial Neural Network Based Nuclei Segmentation on Cytology Pleural Effusion Images. International Conference on Intelligent Informatics and Biomedical Sciences, Okinawa, 24-26 November 2017, 245-249. [8] Krizhevsky, A., Sutskever, I. and Hinton, G.E. (2012) ImageNet Classification with Deep Convolutional Neural Networks. International Conference on Neural Information Pro-cessing Systems, Curran Associates Inc., 1097-1105. [9] Janowczyk, A. and Madabhushi, A. (2016) Deep Learning for Digital Pa-thology Image Analysis: A Comprehensive Tutorial with Selected Use Cases. Journal of Pathology Informatics, 7, 29-47. https://doi.org/10.4103/2153-3539.186902 [10] Macenko, M., Niethammer, M., Marron, J.S., Borland, D., Woosley, J.T., Guan, X., Schmitt, C. and Thomas, N.E. (2009) A Method for Normalizing Histology Slides for Quantitative Analysis. IEEE International Symposium on Biomedical Imaging: From Nano to Macro, Boston, 28 June-1 July 2009, 1107-1110.