On the Explainability of Graph Convolutional Network With GCN Tangent Kernel.

Graph convolutional network (GCN) is a powerful deep model in dealing with graph data. However, the explainability of GCN remains a difficult problem since the training behaviors for graph neural networks are hard to describe. In this work, we show that for GCN with wide hidden feature dimension, the output for semisupervised problem can be described by a simple differential equation. In addition, the dynamics of the behavior of output is decided by the graph convolutional neural tangent kernel (GCNTK), which is stable when the width of hidden feature tends to be infinite. And the solution of node classification can be explained directly by the differential equation for a semisupervised problem. The experiments on some toy models speak to the consistency of the GCNTK model and GCN.

VENet: Variational energy network for gland segmentation of pathological images and early gastric cancer diagnosis of whole slide images.

BACKGROUND AND OBJECTIVE: Gland segmentation of pathological images is an essential but challenging step for adenocarcinoma diagnosis. Although deep learning methods have recently made tremendous progress in gland segmentation, they have not given satisfactory boundary and region segmentation results of adjacent glands. These glands usually have a large difference in glandular appearance, and the statistical distribution between the training and test sets in deep learning is inconsistent. These problems make networks not generalize well in the test dataset, bringing difficulties to gland segmentation and early cancer diagnosis. METHODS: To address these problems, we propose a Variational Energy Network named VENet with a traditional variational energy Lv loss for gland segmentation of pathological images and early gastric cancer detection in whole slide images (WSIs). It effectively integrates the variational mathematical model and the data-adaptability of deep learning methods to balance boundary and region segmentation. Furthermore, it can effectively segment and classify glands in large-size WSIs with reliable nucleus width and nucleus-to-cytoplasm ratio features. RESULTS: The VENet was evaluated on the 2015 MICCAI Gland Segmentation challenge (GlaS) dataset, the Colorectal Adenocarcinoma Glands (CRAG) dataset, and the self-collected Nanfang Hospital dataset. Compared with state-of-the-art methods, our method achieved excellent performance for GlaS Test A (object dice 0.9562, object F1 0.9271, object Hausdorff distance 73.13), GlaS Test B (object dice 94.95, object F1 95.60, object Hausdorff distance 59.63), and CRAG (object dice 95.08, object F1 92.94, object Hausdorff distance 28.01). For the Nanfang Hospital dataset, our method achieved a kappa of 0.78, an accuracy of 0.9, a sensitivity of 0.98, and a specificity of 0.80 on the classification task of test 69 WSIs. CONCLUSIONS: The experimental results show that the proposed model accurately predicts boundaries and outperforms state-of-the-art methods. It can be applied to the early diagnosis of gastric cancer by detecting regions of high-grade gastric intraepithelial neoplasia in WSI, which can assist pathologists in analyzing large WSI and making accurate diagnostic decisions.

Robustness meets accuracy in adversarial training for graph autoencoder.

Graph autoencoder (GAE) is an effective deep method for graph embedding, while it is vulnerable to the graph adversarial attacks. Adversarial training, which generates adversarial examples in the adversarial scope(neighborhood of natural examples), is effective to improve the robustness of GAE. However, it may lead to degradation of natural accuracy (accuracy on natural examples) due to the extra training examples generated in the adversarial scope (the reasonable scope of adversarial examples). Therefore, considering robustness and natural accuracy is crucial to GAE. In this paper, an improved GAE model is formulated by combining the Structure and Feature encoders, and a novel Adversarial Training strategy (GAE-SFAT) is proposed based on improved GAE. GAE-SFAT has a smaller but more reasonable adversarial scope for adversarial training, which keeps the robustness and reduces the degradation of natural accuracy compared with ordinary adversarial training. In addition, a novel algorithm considering the robustness and accuracy is designed to optimize the GAE-SFAT. We conduct experiments both on the natural graphs as well as perturbed graphs for three datasets. The results show that GAE-SFAT can perform better than state of arts adversarial training model under different kinds of perturbations.