Home › Research › Getting a diagnosis
LUMEN: Prototype Conversational AI to Streamline Dementia Assessments
Song Ling Tang, Alexander Robertson, Huizhi Liang, John‐Paul Taylor, Judith Rose Harrison. BJPsych Open, 2025-06-01
Topic: Risk, prevention and care in population studies
LUMEN demonstrated strong performance in distinguishing dementia from normal cognition (AUROC=0.89) but moderate subtype differentiation (AUROC=0.66). Agreement between LUMEN and clinician evaluations was substantial (Cohen’s κ=0.82). However, Lewy body dementia (DLB) identification lagged due to symptom-reporting inaccuracies. System Usability Scale (SUS) scores (mean=82/100) exceeded the ‘excellent’ threshold (≥80). PPPI feedback highlighted LUMEN’s potential to standardise assessment and reduce waiting times. LUMEN is a promising conversational AI tool for improving dementia diagnostics. Gathering caregivers’ collateral input before appointments could streamline workflows within existing outpatient systems and improve clinical accuracy. Real-world trials would help assess workflow integration and mitigate vignette-based biases from simulated testing, such as the overrepresentation of typical phenotypes. This study was conducted in collaboration with Mr Bede Burston, Dr Elizabeth Robertson, and Dr Donncha Mullin, whose contributions were invaluable.
Key findings
- LUMEN demonstrated strong performance in distinguishing dementia from normal cognition (AUROC=0.89) but moderate subtype differentiation (AUROC=0.66).
- Agreement between LUMEN and clinician evaluations was substantial (Cohen’s κ=0.82).
- However, Lewy body dementia (DLB) identification lagged due to symptom-reporting inaccuracies.
- System Usability Scale (SUS) scores (mean=82/100) exceeded the ‘excellent’ threshold (≥80).
- PPPI feedback highlighted LUMEN’s potential to standardise assessment and reduce waiting times.
Why it matters
A prototype conversational AI tool distinguished dementia from normal cognition with strong performance but had only moderate ability to differentiate subtypes. This is a preliminary tool for assessment and does not provide a treatment, cure, or prevention for Alzheimer's disease.
New to this subject? Start with our plain-English guide: Getting a diagnosis →
Study details
- Evidence strengthLow
How this study scored 31/100
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 (BJPsych Open) |
| Journal's citation record | 1 / 4 | BJPsych Open: journal h-index 75 (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 | 4 / 5 | 2025 — within 3 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 | 1 / 7 | Cited 1 times (OpenAlex, 2026-10-10) |
| Citations per year | 0 / 5 | 0.3 citations a year over 3 year(s) since 2025 |
| Impact for its field and year | 1 / 6 | Cited 0.6× the average for papers of its field and year (OpenAlex FWCI) |
| Cited by studies here | 0 / 6 | No other study in this library cites it |
| Still being cited | 0 / 4 | Cited 1 times in 2024–2026 (2026 still in progress; OpenAlex) |
| Senior author's track record | 1 / 4 | Last-listed author Judith Rose Harrison: h-index 10 (OpenAlex, whole career) |
| Institution's track record | 1 / 3 | Newcastle University: institution h-index 932 (OpenAlex) |
| Multi-centre collaboration | 1 / 3 | OpenAlex lists 2 institution(s) in 1 country |
| Indexed in PubMed | 1 / 2 | Not in PubMed, but has a DOI |
| Free to read | 2 / 2 | Free to read |
| Total | 31 / 100 | — |
What each of these 20 components measures, and what this score is not →
Read the original
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
Tang, S. L., Robertson, A., Liang, H., Taylor, J., & Harrison, J. R. (2025). LUMEN: Prototype Conversational AI to Streamline Dementia Assessments. BJPsych Open. https://doi.org/10.1192/bjo.2025.10185
Vancouver
Tang SL, Robertson A, Liang H, Taylor J, Harrison JR. LUMEN: Prototype Conversational AI to Streamline Dementia Assessments. BJPsych Open. 2025. doi:10.1192/bjo.2025.10185.
BibTeX
@article{tang2025-a9f822e1,
author = {Tang, Song Ling and Robertson, Alexander and Liang, Huizhi and Taylor, John‐Paul and Harrison, Judith Rose},
title = {{LUMEN: Prototype Conversational AI to Streamline Dementia Assessments}},
journal = {BJPsych Open},
year = {2025},
doi = {10.1192/bjo.2025.10185},
url = {https://doi.org/10.1192/bjo.2025.10185}
}