Interdisciplinary Expertise to Advance Equitable Explainable AI

TL;DR

This paper proposes an interdisciplinary expert panel review framework to improve the explainability of AI in healthcare, aiming to reduce bias and enhance equity through diverse perspectives and critical assessment.

Contribution

It introduces a novel interdisciplinary review process for XAI that incorporates social epidemiology and health equity expertise to identify bias and guide future research.

Findings

Interdisciplinary panels can identify biases and confounders in AI explanations.

Panel discussions lead to more equitable and contextually informed AI interpretations.

Framework supports targeted improvements in AI models for health equity.

Abstract

The field of artificial intelligence (AI) is rapidly influencing health and healthcare, but bias and poor performance persists for populations who face widespread structural oppression. Previous work has clearly outlined the need for more rigorous attention to data representativeness and model performance to advance equity and reduce bias. However, there is an opportunity to also improve the explainability of AI by leveraging best practices of social epidemiology and health equity to help us develop hypotheses for associations found. In this paper, we focus on explainable AI (XAI) and describe a framework for interdisciplinary expert panel review to discuss and critically assess AI model explanations from multiple perspectives and identify areas of bias and directions for future research. We emphasize the importance of the interdisciplinary expert panel to produce more accurate, equitable interpretations which are historically and contextually informed. Interdisciplinary panel discussions can help reduce bias, identify potential confounders, and identify opportunities for additional research where there are gaps in the literature. In turn, these insights can suggest opportunities for AI model improvement.