A Mixture of Experts (MoE) model to improve AI-based computational pathology prediction performance under variable levels of histopathology image blur

TL;DR

This paper introduces a mixture of experts (MoE) model that combines predictions from multiple models trained on different levels of image blur to improve AI-based pathology predictions on histopathology images with variable quality.

Contribution

The study proposes a novel MoE approach that enhances prediction accuracy in histopathology analysis under image blur, outperforming baseline models across multiple scenarios.

Findings

MoE models outperform baseline models in blurred image scenarios.

Performance decreases with increased blur in baseline models.

MoE-CNN_CLAM and MoE-UNI_CLAM show significant accuracy improvements.

Abstract

AI-based models for histopathology whole slide image (WSI) analysis are increasingly common, but unsharp or blurred areas within WSI can significantly reduce prediction performance. In this study, we investigated the effect of image blur on deep learning models and introduced a mixture of experts (MoE) strategy that combines predictions from multiple expert models trained on data with varying blur levels. Using H&E-stained WSIs from 2,093 breast cancer patients, we benchmarked performance on grade classification and IHC biomarker prediction with both CNN- (CNN_CLAM and MoE-CNN_CLAM) and Vision Transformer-based (UNI_CLAM and MoE-UNI_CLAM) models. Our results show that baseline models' performance consistently decreased with increasing blur, but expert models trained on blurred tiles and especially our proposed MoE approach substantially improved performance, and outperformed baseline models in a range of simulated scenarios. MoE-CNN_CLAM outperformed the baseline CNN_CLAM under moderate (AUC: 0.868 vs. 0.702) and mixed blur conditions (AUC: 0.890 vs. 0.875). MoE-UNI_CLAM outperformed the baseline UNI_CLAM model in both moderate (AUC: 0.950 vs. 0.928) and mixed blur conditions (AUC: 0.944 vs. 0.931). This MoE method has the potential to enhance the reliability of AI-based pathology models under variable image quality, supporting broader application in both research and clinical settings.