Exploring Gender Bias in Alzheimer's Disease Detection: Insights from Mandarin and Greek Speech Perception

2025-07-16

AI Summary

This study identifies, for the first time, a systematic gender bias in auditory perception of Alzheimer’s disease (AD): male voices are significantly more likely to be misclassified as AD-positive than female voices—across both Mandarin Chinese and Greek bilingual listening experiments—indicating language-invariant bias. Method: We combined perceptual evaluation with acoustic analysis, extracting features including fundamental frequency (F0), shimmer, and utterance duration. Contribution/Results: We empirically demonstrate that shimmer positively correlates with AD perception, whereas utterance duration negatively correlates; critically, higher shimmer in male voices partially explains the cross-linguistic gender bias. This is the first evidence of stable, language-independent gender bias in AD voice detection rooted in sex-linked acoustic properties. Our findings underscore that robust AD speech screening models must explicitly model and correct for gender confounding effects—providing key methodological guidance for developing fair and generalizable medical AI systems.

Technology Category

• Natural Language Processing: Ethics — Bias, Fairness, Transparency & Privacy
• Machine Learning: Ethics, Bias, and Fairness
• Computer Vision: Bias, Fairness & Privacy

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: Multilingual and cross-lingual Web search

Abstract

Gender bias has been widely observed in speech perception tasks, influenced by the fundamental voicing differences between genders. This study reveals a gender bias in the perception of Alzheimer's Disease (AD) speech. In a perception experiment involving 16 Chinese listeners evaluating both Chinese and Greek speech, we identified that male speech was more frequently identified as AD, with this bias being particularly pronounced in Chinese speech. Acoustic analysis showed that shimmer values in male speech were significantly associated with AD perception, while speech portion exhibited a significant negative correlation with AD identification. Although language did not have a significant impact on AD perception, our findings underscore the critical role of gender bias in AD speech perception. This work highlights the necessity of addressing gender bias when developing AD detection models and calls for further research to validate model performance across different linguistic contexts.

Problem

• Investigates gender bias in Alzheimer's Disease speech perception
• Examines male speech more frequently identified as AD
• Highlights need to address gender bias in AD detection models

Innovation

• Used perception experiment with Chinese and Greek speech
• Analyzed acoustic features like shimmer and speech portion
• Highlighted gender bias in Alzheimer's Disease detection models