Alzheimer's Disease Detection via Large Pre-Trained Speech and Language Models: Taxonomy, Progress, and Open Challenges

Topic: Brain imaging, connectivity and assessment

Alzheimer's Disease (AD) detection from spoken language has advanced considerably with the rise of large pre-trained speech and language models, including wav2vec 2.0, BERT, generative LLMs, and emerging audio-LLMs. These models often report higher performance on shared AD detection benchmarks than handcrafted-feature pipelines, while reducing the need for manual feature design. Yet their advantages remain difficult to interpret: pre-trained representations are largely a black box, benchmark gains are not always statistically robust, and performance can vary substantially across languages, elicitation protocols, and clinical task definitions.This survey proposes a taxonomy of how large pre-trained speech and language models have been used for AD detection and progression-related tasks. We define Large Pre-trained Models (LPMs) as Transformer-based models trained with self-supervised or weakly supervised objectives, distinguishing them from earlier non-Transformer pre-trained representations. We then organize the literature into six usage strategies: direct application, fine-tuning, transfer learning, multitask learning, multilingual modeling, and generative-LLM-based assessment. Across these categories, we synthesize evidence from shared benchmarks and recent studies to answer four questions:1) what taxonomy captures how LPMs are used for AD detection;2) what progress and advantages do LPMs offer for AD detection;3) what challenges and limitations do LPMs exhibit; and4) what future opportunities and open tasks they present.

@article{li2026-67079121,
  author  = {Li, Y. and Pan, Y. and Liu, Y. and He, L. and Sun, Q. and Zhang, S. and Zhang, W. and Luz, S. and others},
  title   = {{Alzheimer's Disease Detection via Large Pre-Trained Speech and Language Models: Taxonomy, Progress, and Open Challenges}},
  journal = {PsyArXiv},
  year    = {2026},
  doi     = {10.31234/osf.io/s57tz_v2},
  url     = {https://doi.org/10.31234/osf.io/s57tz_v2}
}

Key findings

• These models often report higher performance on shared AD detection benchmarks than handcrafted-feature pipelines, while reducing the need for manual feature design.
• Yet their advantages remain difficult to interpret: pre-trained representations are largely a black box, benchmark gains are not always statistically robust, and performance can vary substantially across languages, elicitation protocols, and clinical task definitions.This survey proposes a taxonomy of how large pre-trained speech and language models have been used for AD detection and progression-related tasks.
• We define Large Pre-trained Models (LPMs) as Transformer-based models trained with self-supervised or weakly supervised objectives, distinguishing them from earlier non-Transformer pre-trained representations.
• We then organize the literature into six usage strategies: direct application, fine-tuning, transfer learning, multitask learning, multilingual modeling, and generative-LLM-based assessment.
• Across these categories, we synthesize evidence from shared benchmarks and recent studies to answer four questions:1) what taxonomy captures how LPMs are used for AD detection;2) what progress and advantages do LPMs offer for AD detection;3) what challenges and limitations do LPMs exhibit; and4) what future opportunities and open tasks they present.

New to this subject? Start with our plain-English guide: Digital health →

Study details

• Evidence strengthLow

Read the original

• DOI: 10.31234/osf.io/s57tz_v2
• Original source →

Free copy unknown: no free-copy check has an answer for this record yet. That does not mean it is paywalled.

Related studies in Digital health tools

• AD-BERT: Using pre-trained language model to predict the progression from mild cognitive impairment to Alzheimer's disease Journal of biomedical informatics · 2023-07-08 · usefulness score 50/100
• Predicting dementia from spontaneous speech using large language models PLOS digital health · 2022-12-22 · usefulness score 46/100
• EEG Pathology Detection via Clinical Concept Bottleneck Models bioRxiv · 2026-10-11 · usefulness score 16/100
• Research progress in predicting the conversion from mild cognitive impairment to Alzheimer’s disease via multimodal MRI and artificial intelligence Frontiers in Neurology · 2025-06-02 · usefulness score 43/100
• Prediction of cognitive impairment via deep learning trained with multi-center neuropsychological test data BMC medical informatics and decision making · 2019-11-21 · usefulness score 45/100
• Prediction of Alzheimer's disease progression within 6 years using speech: A novel approach leveraging language models Alzheimer's & dementia : the journal of the Alzheimer's Association · 2024-06-25 · usefulness score 57/100

Also tagged “convolutional”

205 studies in the library carry the topic tag “convolutional”; these are this study's neighbours in a fixed, alphabetical-by-hash tour of them, not the closest in content. All 205 tagged studies.

• Improving brain age prediction models: incorporation of amyloid status in Alzheimer's disease Neurobiology of aging · 2019-11-14 · usefulness score 39/100 · Risk factors
• Conditional Latent Diffusion for Synthetic Brain MRI in Alzheimer’s Disease: A Preprocessing-Focused Pipeline Preprints.org · 2026-08-07 · usefulness score 16/100 · Digital health tools
• Speech-Based Digital Biomarkers for Alzheimer's Research Methods in molecular biology (Clifton, N.J.) · 2024 · usefulness score 43/100 · Cognitive tests and assessment

Similar studies

• Transformer-based deep neural network language models for Alzheimer's disease risk assessment from targeted speech 2021 · Digital health tools
• nterpretable Alzheimer’s Disease Detection from CHAT Transcripts Using Criterion-Guided LLM Prompting and In-Context Learning 2026 · Digital health tools
• Predicting Cognitive Function Using Transformer-Derived Speech Representations and Longitudinal Coherence Features 2026 · Cognitive tests and assessment
• Detecting Alzheimer’s Disease through Chinese Speech: A Contrast of Manual and Automatic Linguistic Feature Extraction 2026 · Digital health tools
• Small-Sized Reasoning Language Models for Linguistic Screening of Alzheimer’s Disease 2026 · Digital health tools
• Linguistic Features Identify Alzheimer's Disease in Narrative Speech 2016 · Digital health tools

More on Digital health tools

• A review that covers this ground A systematic literature review of automatic Alzheimer's disease detection from speech and language Journal of the American Medical Informatics Association : JAMIA · 2020-11-01 · usefulness score 60/100
• Newest in this subject Speaker-Wise Dementia Screening fromNaturalistic Speech with Lexical Entropy, Semantic Drift, and Multimodal Fusion Research Square · 2026-10-21 · Preprint — not yet peer-reviewed · usefulness score 16/100
• Newest in this subject Integrating Multisource Target Evidence with Machine-Learning Potency Prediction for Alzheimer’s Disease Drug Discovery bioRxiv · 2026-10-20 · Preprint — not yet peer-reviewed · usefulness score 16/100

See all studies in Digital health tools →