Multi-Scale Attention-based Multiple Instance Learning for Classification of Multi-Gigapixel Histology Images

TL;DR

This paper introduces a multi-scale attention-based multiple instance learning approach for classifying large histology images, achieving high accuracy in predicting NPC LMP1 status with interpretability features.

Contribution

It presents the first deep learning method for predicting LMP1 status in NPC histology images using multi-scale attention and MIL.

Findings

Achieved 0.936 accuracy in classification

Achieved 0.995 AUC in prediction

Model interpretability through attention visualization

Abstract

Histology images with multi-gigapixel of resolution yield rich information for cancer diagnosis and prognosis. Most of the time, only slide-level label is available because pixel-wise annotation is labour intensive task. In this paper, we propose a deep learning pipeline for classification in histology images. Using multiple instance learning, we attempt to predict the latent membrane protein 1 (LMP1) status of nasopharyngeal carcinoma (NPC) based on haematoxylin and eosin-stain (H&E) histology images. We utilised attention mechanism with residual connection for our aggregation layers. In our 3-fold cross-validation experiment, we achieved average accuracy, AUC and F1-score 0.936, 0.995 and 0.862, respectively. This method also allows us to examine the model interpretability by visualising attention scores. To the best of our knowledge, this is the first attempt to predict LMP1 status on NPC using deep learning.