Label uncertainty-guided multi-stream model for disease screening

TL;DR

This paper introduces a multi-stream model that leverages label uncertainty to improve disease screening accuracy, especially for challenging cases, by separating simple and hard cases based on uncertainty information.

Contribution

It proposes a novel uncertainty-guided multi-stream network that explicitly handles intra-observer variability in medical image diagnosis.

Findings

Outperforms baseline models in glaucoma screening tasks.

Enhances detection accuracy for hard cases.

Effectively incorporates label uncertainty into model training.

Abstract

The annotation of disease severity for medical image datasets often relies on collaborative decisions from multiple human graders. The intra-observer variability derived from individual differences always persists in this process, yet the influence is often underestimated. In this paper, we cast the intra-observer variability as an uncertainty problem and incorporate the label uncertainty information as guidance into the disease screening model to improve the final decision. The main idea is dividing the images into simple and hard cases by uncertainty information, and then developing a multi-stream network to deal with different cases separately. Particularly, for hard cases, we strengthen the network's capacity in capturing the correct disease features and resisting the interference of uncertainty. Experiments on a fundus image-based glaucoma screening case study show that the proposed model outperforms several baselines, especially in screening hard cases.