基于图卷积神经网络的胃癌和结直肠癌的生存预测
Survival Prediction of Gastric Cancer and Colorectal Cancer Based on Graph Convolutional Neural Networks
摘要: 胃癌和结直肠癌是我国恶性肿瘤中比较常见的两类，同时也是难以治愈的两种癌症。医学界为了统计癌症病人的生存率，提出了5年生存率作为一个有效指标，以此统计癌症病人的存活率。本文将胃癌和结直肠癌的全组织病理学图像(WSI)进行了切片，将切片后的图像进行特征提取后，进行了患者层面的图的构造；将构造好的图形放入构造好的4层图卷积神经网络(GCN)中进行训练，结合每个患者的总生存时间和生存状态，得到了胃癌和结直肠癌的C-index值分别为0.58和0.65，二者结果均高于之前提出的卷积神经网络模型。
Abstract: Gastric cancer and colorectal cancer are two common types of malignant tumors in China, and they are also two cancers that are difficult to cure. In order to calculate the survival rate of cancer pa-tients, the medical community proposed the 5-year survival rate as an effective indicator to calcu-late the survival rate of cancer patients. In this paper, the whole histopathological images (WSI) of gastric cancer and colorectal cancer were sectioned, and the patient-level map was constructed af-ter feature extraction of the sliced images. The constructed graph was put into the constructed 4-layer graph convolutional neural network (GCN) for training, and combined with the total survival time and survival state of each patient, the C-index values of gastric cancer and colorectal cancer were obtained to be 0.58 and 0.65, respectively, which were higher than the previously proposed convolutional neural network model.
文章引用：路晓雯. 基于图卷积神经网络的胃癌和结直肠癌的生存预测[J]. 应用数学进展, 2023, 12(9): 4105-4111. https://doi.org/10.12677/AAM.2023.129402

参考文献

[1] Balkwill, F.R., Capasso, M. and Hagemann, T. (2012) The Tumor Microenvironment at a Glance. Journal of Cell Sci-ence, 125, 5591-5596. [Google Scholar] [CrossRef] [PubMed]
[2] Saltz, J., Gupta, R., Hou, L., Kurc, T., Singh, P., Nguyen, V., Samaras, D., Shroyer, K.R., Zhao, T., Batiste, R., et al. (2018) Spatial Organization and Molecular Correla-tion of Tumor-Infifiltrating Lymphocytes Using Deep Learning on Pathology Images. Cell Reports, 23, 181-193.E7. [Google Scholar] [CrossRef] [PubMed]
[3] Lu, C., Romo-Bucheli, D., Wang, X., Janowczyk, A., Ganesan, S., Gilmore, H., Rimm, D. and Madabhushi, A. (2018) Nuclear Shape and Orientation Features from H&E Images Pre-dict Survival in Early-Stage Estrogen Receptor-Positive Breast Cancers. Laboratory Investigation, 98, 1438-1448. [Google Scholar] [CrossRef] [PubMed]
[4] Zhu, X.L., Yao, J.W., Zhu, F.Y. and Huang, J.Z. (2017) WSISA: Making Survival Prediction from Whole Slide Histopathological Images. 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, 21-26 July 2017, 6855-6863. [Google Scholar] [CrossRef] 
[5] Chen, R.J., Lu, M.Y., Shaban, M., et al. (2021) Whole Slide Images Are 2D Point Clouds: Context-Aware Survival Prediction Using Patch-Based Graph Convolutional Networks. In: de Bruijne, M., et al., Eds., Medical Image Computing and Computer Assisted Intervention—MICCAI 2021, Springer, Cham, 339-349. [Google Scholar] [CrossRef] 
[6] Li, R.Y., Yao, J.W., Zhu, X.L., Li, Y.Q. and Huang, J.Z. (2018) Graph CNN for Survival Analysis on Whole Slide Pathological Images. In: Frangi, A., Schnabel, J., Davatzikos, C., Alberola-López, C. and Fichtinger, G., Eds., Medical Image Computing and Computer Assisted Interven-tion—MICCAI 2018, Springer, Cham, 174-182. [Google Scholar] [CrossRef] 
[7] Harrell Jr., F.E., Lee, K.L. and Mark, D.B. (1996) Multivaria-ble Prognostic Models: Issues in Developing Models, Evaluating Assumptions and Adequacy, and Measuring and Re-ducing Errors. Statistics in Medicine, 15, 361-387. [Google Scholar] [CrossRef] 

为你推荐