Abstract:Background and objectives: Graph neural network (GNN) has been extensively used in histopathology whole slide image (WSI) analysis due to the efficiency and flexibility in modelling relationships among entities. However, most existing GNN-based WSI analysis methods only consider the pairwise correlation of patches from one single perspective (e.g. spatial affinity or embedding similarity) yet ignore the intrinsic non-pairwise relationships present in gigapixel WSI, which are likely to contribute to feature learning and downstream tasks. The objective of this study is therefore to explore the non-pairwise relationships in histopathology WSI and exploit them to guide the learning of slide-level representations for better classification performance. Methods: In this paper, we propose a novel Masked HyperGraph Learning (MaskHGL) framework for weakly supervised histopathology WSI classification. Compared with most GNN-based WSI classification methods, MaskHGL exploits the non-pairwise correlations between patches with hypergraph and global message passing conducted by hypergraph convolution. Concretely, multi-perspective hypergraphs are first built for each WSI, then hypergraph attention is introduced into the jointed hypergraph to propagate the non-pairwise relationships and thus yield more discriminative node representation. More importantly, a masked hypergraph reconstruction module is devised to guide the hypergraph learning which can generate more powerful robustness and generalization than the method only using hypergraph modelling. Additionally, a self-attention-based node aggregator is also applied to explore the global correlation of patches in WSI and produce the slide-level representation for classification. Results: The proposed method is evaluated on two public TCGA benchmark datasets and one in-house dataset. On the public TCGA-LUNG (1494 WSIs) and TCGA-EGFR (696 WSIs) test set, the area under receiver operating characteristic (ROC) curve (AUC) were 0.9752±0.0024 and 0.7421±0.0380, respectively. On the USTC-EGFR (754 WSIs) dataset, MaskHGL achieved significantly better performance with an AUC of 0.8745±0.0100, which surpassed the second-best state-of-the-art method SlideGraph+ 2.64%. Conclusions: MaskHGL shows a great improvement, brought by considering the intrinsic non-pairwise relationships within WSI, in multiple downstream WSI classification tasks. In particular, the designed masked hypergraph reconstruction module promisingly alleviates the data scarcity and greatly enhances the robustness and classification ability of our MaskHGL. Notably, it has shown great potential in cancer subtyping and fine-grained lung cancer gene mutation prediction from hematoxylin and eosin (H&E) stained WSIs.

Topological Feature Extraction and Visualization of Whole Slide Images using Graph Neural Networks

Topological Feature Extraction and Visualization of Whole Slide Images using Graph Neural Networks

Generating Hypergraph-Based High-Order Representations of Whole-Slide Histopathological Images for Survival Prediction

A deep learning based graph-transformer for whole slide image classification

Slide-based Graph Collaborative Training for Histopathology Whole Slide Image Analysis

Histopathology Whole Slide Image Analysis with Heterogeneous Graph Representation Learning

Representation Learning of Histopathology Images using Graph Neural Networks

Graph Neural Networks in Histopathology: Emerging Trends and Future Directions

A graph-transformer for whole slide image classification

Dynamic Graph Representation with Knowledge-aware Attention for Histopathology Whole Slide Image Analysis

Masked hypergraph learning for weakly supervised histopathology whole slide image classification

Fast whole-slide cartography in colon cancer histology using superpixels and CNN classification

Deep learning based registration of serial whole-slide histopathology images in different stains

Derivation of prognostic contextual histopathological features from whole-slide images of tumours via graph deep learning

PATHS: A Hierarchical Transformer for Efficient Whole Slide Image Analysis

HIGT: Hierarchical Interaction Graph-Transformer for Whole Slide Image Analysis

From Histopathology Images to Cell Clouds: Learning Slide Representations with Hierarchical Cell Transformer

Interpretable Histopathology Image Diagnosis Via Whole Tissue Slide Level Supervision

A Deep Learning Approach for Histology-Based Nucleus Segmentation and Tumor Microenvironment Characterization

An Efficient Cervical Whole Slide Image Analysis Framework Based on Multi-scale Semantic and Location Deep Features

Evolutionary Computation in Action: Feature Selection for Deep Embedding Spaces of Gigapixel Pathology Images