![[2602.14655] Breaking Data Efficiency Dilemma: A Federated and Augmented Learning Framework For Alzheimer's Disease Detection via Speech](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2602.14655] Breaking Data Efficiency Dilemma: A Federated and Augmented Learning Framework For Alzheimer's Disease Detection via Speech
Summary
This article presents a novel framework, FAL-AD, that enhances data efficiency in Alzheimer's Disease detection through federated and augmented learning techniques, achieving state-of-the-art accuracy.
Why It Matters
The research addresses the critical challenge of data scarcity and privacy in medical AI applications. By integrating federated learning with data augmentation, it offers a practical solution for improving diagnostic accuracy in Alzheimer's Disease detection, which is vital for early intervention.
Key Takeaways
- FAL-AD framework improves data efficiency for Alzheimer's detection.
- Utilizes voice conversion-based augmentation for diverse speech samples.
- Implements adaptive federated learning to maximize cross-institutional benefits.
- Achieves a multi-modal accuracy of 91.52%, outperforming centralized methods.
- Source code is publicly available, promoting further research and application.
Computer Science > Computation and Language arXiv:2602.14655 (cs) [Submitted on 16 Feb 2026] Title:Breaking Data Efficiency Dilemma: A Federated and Augmented Learning Framework For Alzheimer's Disease Detection via Speech Authors:Xiao Wei, Bin Wen, Yuqin Lin, Kai Li, Mingyang gu, Xiaobao Wang, Longbiao Wang, Jianwu Dang View a PDF of the paper titled Breaking Data Efficiency Dilemma: A Federated and Augmented Learning Framework For Alzheimer's Disease Detection via Speech, by Xiao Wei and 7 other authors View PDF HTML (experimental) Abstract:Early diagnosis of Alzheimer's Disease (AD) is crucial for delaying its progression. While AI-based speech detection is non-invasive and cost-effective, it faces a critical data efficiency dilemma due to medical data scarcity and privacy barriers. Therefore, we propose FAL-AD, a novel framework that synergistically integrates federated learning with data augmentation to systematically optimize data efficiency. Our approach delivers three key breakthroughs: First, absolute efficiency improvement through voice conversion-based augmentation, which generates diverse pathological speech samples via cross-category voice-content recombination. Second, collaborative efficiency breakthrough via an adaptive federated learning paradigm, maximizing cross-institutional benefits under privacy constraints. Finally, representational efficiency optimization by an attentive cross-modal fusion model, which achieves fine-grained word-level alignment and ...
