Multi-Stain Self-Attention Graph Multiple Instance Learning Pipeline for Histopathology Whole Slide Images
Amaya Gallagher-Syed (Queen Mary University of London),* Luca Rossi (The Hong Kong Polytechnic University), Felice Rivellese (Queen Mary University of London), Costantino Pitzalis (Queen Mary University of London), Myles Lewis (Queen Mary University of London), Michael Barnes (Queen Mary University of London), Gregory Slabaugh (Queen Mary University of London)
The 34th British Machine Vision Conference
Abstract
Whole Slide Images (WSIs) analysis presents a challenging computer vision task due to their gigapixel size and presence of numerous artefacts. Yet they are a valuable resource for patient diagnosis and stratification, often representing the gold standard for diagnostic tasks. Real-world clinical datasets tend to come as sets of heterogeneous WSIs with labels present at the patient-level, with poor to no annotations. Weakly supervised attention-based multiple instance learning approaches have been developed in recent years to address these challenges, but can fail to resolve both long and short-range dependencies. Here we propose an end-to-end multi-stain self-attention graph (MUSTANG) multiple instance learning pipeline, which is designed to solve a weakly-supervised gigapixel multi-image classification task, where the label is assigned at the patient-level, but no slide-level labels or region annotations are available. The pipeline uses a self-attention based approach by restricting the operations to a highly sparse k-Nearest Neighbour Graph of embedded WSI patches based on the Euclidean distance. We show this approach achieves a state-of-the-art F1-score/AUC of 0.89/0.92, outperforming the widely used CLAM model. Our approach is highly modular and can easily be modified to suit different clinical datasets, as it only requires a patient-level label without annotations and accepts WSI sets of different sizes, as the graphs can be of varying sizes and structures. The source code can be found at https://github.com/AmayaGS/MUSTANG.Video
Citation
@inproceedings{Gallagher-Syed_2023_BMVC,
author = {Amaya Gallagher-Syed and Luca Rossi and Felice Rivellese and Costantino Pitzalis and Myles Lewis and Michael Barnes and Gregory Slabaugh},
title = {Multi-Stain Self-Attention Graph Multiple Instance Learning Pipeline for Histopathology Whole Slide Images},
booktitle = {34th British Machine Vision Conference 2023, {BMVC} 2023, Aberdeen, UK, November 20-24, 2023},
publisher = {BMVA},
year = {2023},
url = {https://papers.bmvc2023.org/0789.pdf}
}
Copyright © 2023
The British Machine Vision Conference is organised by