Adaptive Multi-Scale Graph Transformer Framework Forhistopathological Images
Wu, C.-i.; Banda, K.; Swisher, E.; Sailem, H.
Show abstract
Whole slide images (WSIs) contain hierarchical information from cellular to tissue architecture but their gigapixel scale poses major memory and computational challenges. Existing multi-scale graph and transformer models capture complex WSI features effectively but struggle with efficiency. We propose an Adaptive Multi-Scale Graph Transformer (AMGT) for WSI classification that addresses this limitation through two key modules: a Self-Guided Token Aggregation (SGTA) mechanism that fuses multi-resolution features to reduce redundancy, and a Prototypical Transformer (PT) that groups similar tokens into phenotype-representative prototypes with linear complexity. This design preserves essential spatial and semantic information, substantially lowering memory cost and improving interpretability by prototypical learning. AMGT achieves superior performance and efficiency, outperforming state-of-the-art models by 1.8% and 5.3% AUC on high-grade ovarian cancer and Camelyon16 datasets, respectively. These results demonstrate AMGTs capacity for scalable, interpretable multi-scale representation learning.
Matching journals
The top 8 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Predicting Spatially Resolved Gene Expression via Tissue Morphology using Adaptive Spatial GNNs 94%
- Graspot: A graph attention network for spatial transcriptomics data integration with optimal transport 94%
- A Spatial Attention Guided Deep Learning System for Prediction of Pathological Complete Response Using Breast Cancer Histopathology Images 93%
Similar papers in this journal
Similar papers in this journal
- pathCLIP: Detection of Genes and Gene Relations from Biological Pathway Figures through Image-Text Contrastive Learning 94%
- SimSearch: A Human-in-the-Loop Learning Framework for Fast Detection of Regions of Interest in Microscopy Images 94%
- Dual-Field Microvascular Segmentation: Hemodynamically-Consistent Attention Learning for Retinal Vasculature Mapping 93%
Similar papers in this journal
- Aggregation of Cohorts for Histopathological Diagnosis with Deep Morphological Analysis 95%
- Spatial Transcriptomics Inferred from Pathology Whole-Slide Images Links Tumor Heterogeneity to Survival in Breast and Lung Cancer 95%
- Assisting Scalable Diagnosis Automatically via CT Images in the Combat against COVID-19 94%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.