The gold standard for gastric cancer detection is gastric histopathological image analysis, but there are certain drawbacks in the existing histopathological detection and diagnosis. In this paper, based on the study of computer aided diagnosis system, graph based features are applied to gastric cancer histopathology microscopic image analysis, and a classifier is used to classify gastric cancer cells from benign cells. Firstly, image segmentation is performed, and after finding the region, cell nuclei are extracted using the k-means method, the minimum spanning tree (MST) is drawn, and graph based features of the MST are extracted. The graph based features are then put into the classifier for classification. In this study, different segmentation methods are compared in the tissue segmentation stage, among which are Level-Set, Otsu thresholding, watershed, SegNet, U-Net and Trans-U-Net segmentation; Graph based features, Red, Green, Blue features, Grey-Level Co-occurrence Matrix features, Histograms of Oriented Gradient features and Local Binary Patterns features are compared in the feature extraction stage; Radial Basis Function (RBF) Support Vector Machine (SVM), Linear SVM, Artificial Neural Network, Random Forests, k-NearestNeighbor, VGG16, and Inception-V3 are compared in the classifier stage. It is found that using U-Net to segment tissue areas, then extracting graph based features, and finally using RBF SVM classifier gives the optimal results with 94.29%.
翻译:胃癌检测的金标准是胃癌病理图象分析,但在现有的组织病理学检测和诊断中存在某些缺陷。在本文中,基于计算机辅助诊断系统研究的图表特征应用到胃癌病理组织病理微生物图象分析,并使用分类器对良细胞的胃癌细胞进行分类。首先,进行图像分解,并在发现区域后,用K值方法抽取细胞核,绘制最小覆盖树,并提取MST的图形特征。然后将基于图形的特征置于分类器中。在这一研究中,对组织分解阶段应用了不同的分解方法,其中包括:Set、Otsu临界、流域、SegNet、U-Net和 Trans-U-Net分解;基于图表的特征、红、绿色、蓝、灰度的共振动矩阵特征、Orental-NST的分布图示图,然后将基于图解的图解分解分解特性和本地BIF3 结构结构图解,然后将S-RGF-RF-RF-S-S Frial-S-Ial-S-Simal-Slaimal-Slaimal-Sal-Syal-Slaudal-S-Syal-Syal-Syal-Syal-Sl 和Syal-Smaisal-Sy-Sma-Sma-S-S-S-S-S-S-S-S-S-S-Sal-Sal-Sal-Sal-Sal-Sal-Syal-Syal-Ima-Smax-S-S-S-F-Syal-Sma-Sal-Sal-S-S-S-S-Sal-S-S-S-S-S-S-S-I-S-S-S-S-S-S-S-S-S-S-S-S-Ial-S-S-S-I-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-