What is Fair? Defining Fairness in Machine Learning for Health

2024-06-13

AI Summary

This study addresses the exacerbation of health inequities due to the lack of rigorous, context-sensitive fairness definitions in medical AI. Through a systematic literature review and cross-paradigmatic conceptual analysis—integrating statistical fairness, individual fairness, and counterfactual causal fairness—we propose the first comprehensive fairness taxonomy for medical machine learning, spanning group-level, individual-level, and causal dimensions. We further introduce a novel fairness definition framework that balances clinical interpretability with operational feasibility. Additionally, we rigorously delineate the applicability boundaries and clinical limitations of twelve widely adopted fairness metrics. The resulting framework provides both theoretical foundations and actionable guidance for regulatory evaluation, clinical deployment decisions, and algorithmic governance of medical AI systems. By bridging technical fairness theory with clinical practice, this work advances the development and implementation of equitable, safe, and effective AI in healthcare.

Technology Category

• Philosophy and Ethics of AI: Bias, Fairness & Equity
• Machine Learning: Ethics, Bias, and Fairness
• Humans and AI: Explainable AI (XAI) for Human Understanding

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Social Networks and Social Media: Fairness and bias in social network and social media analysis
• Economics, Online Markets and Human Computation: Fairness and ethical considerations in crowd work and in human-in-the-loop AI systems

Abstract

Ensuring that machine learning (ML) models are safe, effective, and equitable across all patients is critical for clinical decision-making and for preventing the amplification of existing health disparities. In this work, we examine how fairness is conceptualized in ML for health, including why ML models may lead to unfair decisions and how fairness has been measured in diverse real-world applications. We review commonly used fairness notions within group, individual, and causal-based frameworks. We also discuss the outlook for future research and highlight opportunities and challenges in operationalizing fairness in health-focused applications.

Problem

• Defining fairness in ML for health applications
• Examining causes of unfair ML decisions in healthcare
• Reviewing fairness metrics in group and individual contexts

Innovation

• Examining fairness in ML for health applications
• Reviewing group, individual, causal fairness frameworks
• Discussing challenges in operationalizing health fairness