Nuclei classification is a critical step in computer-aided diagnosis with histopathology images. In the past, various methods have employed graph neural networks (GNN) to analyze cell graphs that model inter-cell relationships by considering nuclei as vertices. However, they are limited by the GNN mechanism that only passes messages among local nodes via fixed edges. To address the issue, we develop a cell graph transformer (CGT) that treats nodes and edges as input tokens to enable learnable adjacency and information exchange among all nodes. Nevertheless, training the transformer with a cell graph presents another challenge. Poorly initialized features can lead to noisy self-attention scores and inferior convergence, particularly when processing the cell graphs with numerous connections. Thus, we further propose a novel topology-aware pretraining method that leverages a graph convolutional network (GCN) to learn a feature extractor. The pre-trained features may suppress unreasonable correlations and hence ease the finetuning of CGT. Experimental results suggest that the proposed cell graph transformer with topology-aware pretraining significantly improves the nuclei classification results, and achieves the state-of-the-art performance. Code and models are available at https://github.com/lhaof/CGT

翻译：细胞核分类是基于组织病理学图像的计算机辅助诊断中的关键步骤。以往，多种方法采用图神经网络分析细胞图，通过将细胞核视为顶点来建模细胞间关系。然而，这些方法受限于GNN机制，仅通过固定边在局部节点间传递信息。为克服这一局限，我们开发了一种细胞图Transformer（CGT），将节点和边作为输入标记，从而实现可学习的邻接关系以及所有节点间的信息交换。尽管如此，使用细胞图训练Transformer仍面临挑战。初始化特征不佳可能导致自注意力分数混乱且收敛性差，尤其在处理具有大量连接的细胞图时。因此，我们进一步提出一种新颖的拓扑感知预训练方法，利用图卷积网络学习特征提取器。预训练特征可抑制不合理关联，从而简化CGT的微调过程。实验结果表明，结合拓扑感知预训练的细胞图Transformer显著提升了细胞核分类性能，并达到了当前最优水平。代码和模型可从https://github.com/lhaof/CGT获取。