Enhancing the Reliability of Medical AI through Expert-guided Uncertainty Modeling

TL;DR

This paper introduces a novel expert-guided uncertainty modeling approach for medical AI, improving the estimation of data ambiguity and noise to enhance safety and reliability in healthcare diagnostics.

Contribution

It proposes a disagreement-based training method to separately estimate aleatoric and epistemic uncertainties, validated across multiple medical AI tasks.

Findings

Uncertainty estimation improved by 9% to 50% across tasks.

Expert disagreement effectively guides uncertainty modeling.

Enhanced risk-aware decision-making in healthcare AI systems.

Abstract

Artificial intelligence (AI) systems accelerate medical workflows and improve diagnostic accuracy in healthcare, serving as second-opinion systems. However, the unpredictability of AI errors poses a significant challenge, particularly in healthcare contexts, where mistakes can have severe consequences. A widely adopted safeguard is to pair predictions with uncertainty estimation, enabling human experts to focus on high-risk cases while streamlining routine verification. Current uncertainty estimation methods, however, remain limited, particularly in quantifying aleatoric uncertainty, which arises from data ambiguity and noise. To address this, we propose a novel approach that leverages disagreement in expert responses to generate targets for training machine learning models. These targets are used in conjunction with standard data labels to estimate two components of uncertainty separately, as given by the law of total variance, via a two-ensemble approach, as well as its lightweight variant. We validate our method on binary image classification, binary and multi-class image segmentation, and multiple-choice question answering. Our experiments demonstrate that incorporating expert knowledge can enhance uncertainty estimation quality by $9\%$ to $50\%$ depending on the task, making this source of information invaluable for the construction of risk-aware AI systems in healthcare applications.