ResGAT: A Residual Graph Attention Network for Cancer Subtype Classification in Whole Slide Images

Zhenhan Lin, Hao Tong, Yunfei Hu, Xianyong Sean Gui, Jeanne Shen, Byrne Lee, Lu Zhang, Daniel Moyer, Mu Zhou, Xin Maizie Zhou, Konstantinos Votanopoulos

Keywords: whole slide image classification, multiple instance learning, residual graph attention framework, cross-site generalization

Abstract: Multiple instance learning (MIL) provides a weakly supervised framework for whole slide image (WSI) classification, enabling slide-level prediction from gigapixel images with only slide-level labels. However, WSI subtype classification in realistic settings is still challenging. In this work, we propose ResGAT, a residual graph attention framework that operates on patch graphs and models representations with stacked residual graph attention blocks. ResGAT is evaluated on binary subtype classification task across a rare, class-imbalanced appendiceal cancer cohort and two public TCGA datasets. It outperforms SOTA MIL baselines on the appendiceal cancer cohort and remains competitive on the TCGA datasets. We further assess cross-site generalization via few-shot adaptation under source shift, showing that ResGAT adapts effectively to new domains with limited labels. An ablation study is provided to assess the effectiveness of key architectural components of our method.

Primary Subject Area: Application: Histopathology

Secondary Subject Area: Learning with Noisy Labels and Limited Data

Registration Requirement: Yes

Visa & Travel: Yes

Read CFP & Author Instructions: Yes

Originality Policy: Yes

Single-blind & Not Under Review Elsewhere: Yes

LLM Policy: Yes

Submission Number: 369