AI Alignment in Medical Imaging: Unveiling Hidden Biases Through Counterfactual Analysis

2025-04-28

AI Summary

Medical imaging AI models often exhibit bias with respect to demographic attributes (e.g., sex, race), undermining cross-population generalizability and clinical safety. To address this, we propose the first statistical evaluation framework grounded in counterfactual invariance—requiring no ground-truth counterfactual data—to quantify implicit model dependence on sensitive attributes. Our method innovatively integrates conditional latent diffusion models to generate controllable counterfactual samples, coupled with hypothesis testing for interpretable, statistically rigorous bias quantification. We validate the framework on multi-source chest X-ray datasets (CheXpert and MIMIC-CXR), demonstrating significant improvements over existing baselines in both bias detection and mitigation, while enhancing model robustness. The implementation is publicly available.

Technology Category

• Computer Vision: Bias, Fairness & Privacy
• Philosophy and Ethics of AI: Bias, Fairness & Equity
• Machine Learning: Ethics, Bias, and Fairness

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Social Networks and Social Media: Fairness and bias in social network and social media analysis

Abstract

Machine learning (ML) systems for medical imaging have demonstrated remarkable diagnostic capabilities, but their susceptibility to biases poses significant risks, since biases may negatively impact generalization performance. In this paper, we introduce a novel statistical framework to evaluate the dependency of medical imaging ML models on sensitive attributes, such as demographics. Our method leverages the concept of counterfactual invariance, measuring the extent to which a model's predictions remain unchanged under hypothetical changes to sensitive attributes. We present a practical algorithm that combines conditional latent diffusion models with statistical hypothesis testing to identify and quantify such biases without requiring direct access to counterfactual data. Through experiments on synthetic datasets and large-scale real-world medical imaging datasets, including extsc{cheXpert} and MIMIC-CXR, we demonstrate that our approach aligns closely with counterfactual fairness principles and outperforms standard baselines. This work provides a robust tool to ensure that ML diagnostic systems generalize well, e.g., across demographic groups, offering a critical step towards AI safety in healthcare. Code: https://github.com/Neferpitou3871/AI-Alignment-Medical-Imaging.

Problem

• Detect biases in medical imaging ML models
• Assess model dependency on sensitive attributes
• Ensure fairness across demographic groups

Innovation

• Counterfactual invariance framework for bias evaluation
• Conditional latent diffusion models for bias identification
• Statistical hypothesis testing for bias quantification