FALFormer: Feature-aware Landmarks self-attention for Whole-slide Image Classification

TL;DR

FALFormer is a novel transformer-based model that processes entire whole-slide images using feature-aware landmarks and Nyström self-attention to improve classification accuracy in digital pathology.

Contribution

The paper introduces FALFormer, a new slide-level classification model that fully exploits patch relationships in WSIs using efficient Nyström self-attention and feature-aware landmarks.

Findings

Outperforms state-of-the-art methods on CAMELYON16 and TCGA-BRCA datasets.

Effectively captures global relationships among patches in WSIs.

Demonstrates superior accuracy in digital pathology classification tasks.

Abstract

Slide-level classification for whole-slide images (WSIs) has been widely recognized as a crucial problem in digital and computational pathology. Current approaches commonly consider WSIs as a bag of cropped patches and process them via multiple instance learning due to the large number of patches, which cannot fully explore the relationship among patches; in other words, the global information cannot be fully incorporated into decision making. Herein, we propose an efficient and effective slide-level classification model, named as FALFormer, that can process a WSI as a whole so as to fully exploit the relationship among the entire patches and to improve the classification performance. FALFormer is built based upon Transformers and self-attention mechanism. To lessen the computational burden of the original self-attention mechanism and to process the entire patches together in a WSI, FALFormer employs Nystr\"om self-attention which approximates the computation by using a smaller number of tokens or landmarks. For effective learning, FALFormer introduces feature-aware landmarks to enhance the representation power of the landmarks and the quality of the approximation. We systematically evaluate the performance of FALFormer using two public datasets, including CAMELYON16 and TCGA-BRCA. The experimental results demonstrate that FALFormer achieves superior performance on both datasets, outperforming the state-of-the-art methods for the slide-level classification. This suggests that FALFormer can facilitate an accurate and precise analysis of WSIs, potentially leading to improved diagnosis and prognosis on WSIs.