Invisible Attributes, Visible Biases: Exploring Demographic Shortcuts in MRI-based Alzheimer's Disease Classification

2025-09-11

AI Summary

This study identifies pervasive shortcut learning and demographic bias in MRI-based deep learning models for Alzheimer’s disease diagnosis: models exploit protected attributes—such as race and sex—as spurious features, degrading generalization to underrepresented groups. Through systematic experiments with ResNet and Swin Transformer on multi-center 3D MRI datasets, augmented by quantitative attribution and visualization analyses, we provide the first empirical evidence that latent demographic distribution shifts in neuroimaging data are captured by models and induce bias. We propose the first fairness analysis framework tailored to neuroimaging AI, which explicitly localizes brain regions vulnerable to demographic bias and demonstrates that imbalanced demographic composition in training data significantly compromises model fairness. To foster reproducible, equitable AI research, we release all code and analysis tools publicly.

Technology Category

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

Application Category

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

Abstract

Magnetic resonance imaging (MRI) is the gold standard for brain imaging. Deep learning (DL) algorithms have been proposed to aid in the diagnosis of diseases such as Alzheimer's disease (AD) from MRI scans. However, DL algorithms can suffer from shortcut learning, in which spurious features, not directly related to the output label, are used for prediction. When these features are related to protected attributes, they can lead to performance bias against underrepresented protected groups, such as those defined by race and sex. In this work, we explore the potential for shortcut learning and demographic bias in DL based AD diagnosis from MRI. We first investigate if DL algorithms can identify race or sex from 3D brain MRI scans to establish the presence or otherwise of race and sex based distributional shifts. Next, we investigate whether training set imbalance by race or sex can cause a drop in model performance, indicating shortcut learning and bias. Finally, we conduct a quantitative and qualitative analysis of feature attributions in different brain regions for both the protected attribute and AD classification tasks. Through these experiments, and using multiple datasets and DL models (ResNet and SwinTransformer), we demonstrate the existence of both race and sex based shortcut learning and bias in DL based AD classification. Our work lays the foundation for fairer DL diagnostic tools in brain MRI. The code is provided at https://github.com/acharaakshit/ShortMR

Problem

• Detecting demographic bias in Alzheimer's MRI classification models
• Investigating shortcut learning from race and sex attributes
• Analyzing feature attribution disparities across protected groups

Innovation

• Detects demographic shortcuts in MRI scans
• Analyzes feature attributions across brain regions
• Uses multiple datasets and deep learning models