Home › Research › Digital health tools
Predicting dementia from spontaneous speech using large language models
Agbavor F, Liang H.. PLOS digital health, 2022-12-22
Topic: Risk, prevention and care in population studies
Language impairment is an important biomarker of neurodegenerative disorders such as Alzheimer's disease (AD). Artificial intelligence (AI), particularly natural language processing (NLP), has recently been increasingly used for early prediction of AD through speech. Yet, relatively few studies exist on using large language models, especially GPT-3, to aid in the early diagnosis of dementia. In this work, we show for the first time that GPT-3 can be utilized to predict dementia from spontaneous speech. Specifically, we leverage the vast semantic knowledge encoded in the GPT-3 model to generate text embedding, a vector representation of the transcribed text from speech, that captures the semantic meaning of the input. We demonstrate that the text embedding can be reliably used to (1) distinguish individuals with AD from healthy controls, and (2) infer the subject's cognitive testing score, both solely based on speech data. We further show that text embedding considerably outperforms the conventional acoustic feature-based approach and even performs competitively with prevailing fine-tuned models. Together, our results suggest that GPT-3 based text embedding is a viable approach for AD assessment directly from speech and has the potential to improve early diagnosis of dementia.
Key findings
- Together, our results suggest that GPT-3 based text embedding is a viable approach for AD assessment directly from speech and has the potential to improve early diagnosis of dementia.
Why it matters
This study showed that a large language model could use transcribed speech to distinguish people with Alzheimer's disease from healthy controls and estimate cognitive test scores. It is a diagnostic approach and does not test any treatment, cure, or prevention.
New to this subject? Start with our plain-English guide: Digital health →
Study details
- Evidence strengthLow
How this study scored 46/100
Published in a peer-reviewed journal. Scored zero for study design, sample size, effect reported, studied in people and randomised / blinded because the record does not state it.
| Component | Points | Why |
|---|---|---|
| Journal tier | 7 / 10 | Tier 3 — peer-reviewed journal (PLOS digital health) |
| Journal's citation record | 0 / 4 | PLOS Digital Health: journal h-index 54 (OpenAlex, 2026-10-29) |
| Study design | 0 / 12 | No study design recorded — scores zero, not guessed (not stated in the record — scored 0) |
| Strength of evidence | 1 / 5 | Rated “low” in our summary |
| Peer-review validation | 10 / 10 | Peer-reviewed. Published as an article in a named peer-reviewed journal. |
| How recent | 3 / 5 | 2022 — within 6 year(s) of the newest study here |
| Sample size | 0 / 4 | Participant count not stated — scores zero (not stated in the record — scored 0) |
| Effect reported | 0 / 3 | No effect size reported — scores zero (not stated in the record — scored 0) |
| Studied in people | 0 / 3 | Not stated whether this was studied in people — scores zero (not stated in the record — scored 0) |
| Randomised / blinded | 0 / 2 | No randomisation or blinding stated — scores zero (not stated in the record — scored 0) |
| Times cited | 5 / 7 | Cited 142 times (OpenAlex, 2026-10-10) |
| Citations per year | 4 / 5 | 23.7 citations a year over 6 year(s) since 2022 |
| Impact for its field and year | 4 / 6 | Cited 15.9× the average for papers of its field and year (OpenAlex FWCI) |
| Cited by studies here | 2 / 6 | Cited by 5 other studies in this library |
| Still being cited | 3 / 4 | Cited 123 times in 2024–2026 (2026 still in progress; OpenAlex) |
| Senior author's track record | 2 / 4 | Last-listed author Hualou Liang: h-index 31 (OpenAlex, whole career) |
| Institution's track record | 1 / 3 | Drexel University: institution h-index 761 (OpenAlex) |
| Multi-centre collaboration | 0 / 3 | OpenAlex lists 1 institution(s) in 1 country — scores zero (affiliation lists can be incomplete) |
| Indexed in PubMed | 2 / 2 | Indexed in PubMed (PMID 36812634) |
| Free to read | 2 / 2 | Free to read |
| Total | 46 / 100 | — |
What each of these 20 components measures, and what this score is not →
Read the original
- DOI: 10.1371/journal.pdig.0000168
- PubMed: 36812634
- Original source →
- Free full text: PLOS Digital Health →
Full text: free to read — OpenAlex lists this copy as free to read: the link above (checked 6 Feb 1020).
Cite this study APA, Vancouver, BibTeX · download .bib or .ris
Built from the record alzhub holds for this paper, not looked up. Volume, issue and page numbers are not held in the record. Check the citation against the original before you use it.
APA 7th
Agbavor, F., & Liang, H. (2022). Predicting dementia from spontaneous speech using large language models. PLOS digital health. https://doi.org/10.1371/journal.pdig.0000168
Vancouver
Agbavor F, Liang H. Predicting dementia from spontaneous speech using large language models. PLOS digital health. 2022. doi:10.1371/journal.pdig.0000168. PMID: 36812634.
BibTeX
@article{agbavor2022-36812634,
author = {Agbavor, F. and Liang, H.},
title = {{Predicting dementia from spontaneous speech using large language models}},
journal = {PLOS digital health},
year = {2022},
doi = {10.1371/journal.pdig.0000168},
pmid = {36812634},
url = {https://doi.org/10.1371/journal.pdig.0000168}
}
Studies this one cites
Referenced by this study and also in the alzhub library, so you can read them here.
Later studies that cite this one
5 studies in the alzhub library list this one in their references.