Asymmetric Co-Training with Explainable Cell Graph Ensembling for Histopathological Image Classification

TL;DR

This paper introduces an asymmetric co-training framework combining deep graph convolutional networks and CNNs to improve multi-class histopathological image classification, enhancing explainability and leveraging cell and pixel-level features.

Contribution

The paper proposes a novel asymmetric co-training approach with a deep GCN and CNN, improving classification accuracy and explainability in histopathological images.

Findings

Superior performance on LUAD7C and colorectal datasets

Enhanced explainability through morphological and topological features

Effective integration of pixel and cell-level information

Abstract

Convolutional neural networks excel in histopathological image classification, yet their pixel-level focus hampers explainability. Conversely, emerging graph convolutional networks spotlight cell-level features and medical implications. However, limited by their shallowness and suboptimal use of high-dimensional pixel data, GCNs underperform in multi-class histopathological image classification. To make full use of pixel-level and cell-level features dynamically, we propose an asymmetric co-training framework combining a deep graph convolutional network and a convolutional neural network for multi-class histopathological image classification. To improve the explainability of the entire framework by embedding morphological and topological distribution of cells, we build a 14-layer deep graph convolutional network to handle cell graph data. For the further utilization and dynamic interactions between pixel-level and cell-level information, we also design a co-training strategy to integrate the two asymmetric branches. Notably, we collect a private clinically acquired dataset termed LUAD7C, including seven subtypes of lung adenocarcinoma, which is rare and more challenging. We evaluated our approach on the private LUAD7C and public colorectal cancer datasets, showcasing its superior performance, explainability, and generalizability in multi-class histopathological image classification.