Home › Research › Digital health tools
A retrieval-augmented generation large language model framework for accurate dementia identification from electronic health records
Wang L, Liu B, Yang R, Chuang Y, Estiri H, Murphy S, Zhou L, Marshall GA.. medRxiv, 2026-01-25
Topic: Brain imaging, connectivity and assessment
Results The RAG-based classifier achieved the highest performance (F1=0.933, sensitivity=91.1%, PPV=95.5%) compared to rule-based (F1=0.823, sensitivity=81.1%, PPV=83.5%) and keyword-filtered LLM (F1=0.903, sensitivity=91.7%, PPV=88.6%). Including ICD codes alongside free text in the RAG-based LLM pipeline significantly reduced the PPV and modestly decreased F-1 score. Error analysis revealed that structured-code dependence contributed to false positives, whereas unrecognized contextual cues in notes drove false negatives. Conclusion A RAG-based LLM pipeline without structured ICD codes improved dementia ascertainment from EHR data compared with ICD-based rules and keyword-based filtering.
Key findings
- Results The RAG-based classifier achieved the highest performance (F1=0.933, sensitivity=91.1%, PPV=95.5%) compared to rule-based (F1=0.823, sensitivity=81.1%, PPV=83.5%) and keyword-filtered LLM (F1=0.903, sensitivity=91.7%, PPV=88.6%).
- Including ICD codes alongside free text in the RAG-based LLM pipeline significantly reduced the PPV and modestly decreased F-1 score.
- Error analysis revealed that structured-code dependence contributed to false positives, whereas unrecognized contextual cues in notes drove false negatives.
- Conclusion A RAG-based LLM pipeline without structured ICD codes improved dementia ascertainment from EHR data compared with ICD-based rules and keyword-based filtering.
New to this subject? Start with our plain-English guide: Digital health →
Study details
- Effect size91.1%; 95.5%; 81.1%
- Evidence strengthModerate
How this study scored 16/100
Preprint — not yet peer-reviewed. It scored 27 on the same rubric, but non-peer-reviewed sources are capped at 16, so it shows as 16. Scored zero for study design, sample size, studied in people and randomised / blinded because the record does not state it. We have no data on times cited, citations per year, impact for its field and year, cited by studies here, still being cited, senior author's track record, institution's track record and multi-centre collaboration, so those were left out: it earned 17 of the 62 points we could judge it on, which is 27 out of 100.
| Component | Points | Why |
|---|---|---|
| Journal tier | 2 / 10 | Tier 4 — not a peer-reviewed journal (medRxiv) |
| Journal's citation record | 0 / 4 | Not a journal article, so the journal's record does not count |
| Study design | 0 / 12 | No study design recorded — scores zero, not guessed (not stated in the record — scored 0) |
| Strength of evidence | 3 / 5 | Rated “moderate” in our summary |
| Peer-review validation | 4 / 10 | Preprint — not yet peer-reviewed. Hosted on a preprint server. |
| How recent | 5 / 5 | 2026 — within 1 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 | 2 / 3 | Effect reported as a number, without an interval or test |
| 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 | — / 7 | No citation count (not found in OpenAlex) — left out of this score |
| Citations per year | — / 5 | No citation count or no year — left out of this score |
| Impact for its field and year | — / 6 | No field-weighted citation impact (OpenAlex has none yet) — left out of this score |
| Cited by studies here | — / 6 | Not yet checked against this library's citation graph — left out of this score |
| Still being cited | — / 4 | No yearly citation counts — left out of this score |
| Senior author's track record | — / 4 | Senior author's record not available — left out of this score |
| Institution's track record | — / 3 | No institution listed in OpenAlex — left out of this score |
| Multi-centre collaboration | — / 3 | Institutions not listed — left out of this score |
| Indexed in PubMed | 1 / 2 | Not in PubMed, but has a DOI |
| Free to read | 0 / 2 | Behind a paywall — scores zero |
| Points | 17 / 62 | Scored on 12 of 20 parts. We have no data for the rest, so they are left out: 17 of the 62 points we could judge it on is 27 out of 100. |
| Total | 27 / 100 | Capped to 16 because this is not a peer-reviewed article. |
What each of these 20 components measures, and what this score is not →
Read the original
Free copy unknown: no free-copy check has an answer for this record yet. That does not mean it is paywalled.
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. The record names 8 authors; APA and Vancouver show the first six, BibTeX and the downloads carry all of them. Check the citation against the original before you use it.
APA 7th
Wang, L., Liu, B., Yang, R., Chuang, Y., Estiri, H., Murphy, S., et al. (2026). A retrieval-augmented generation large language model framework for accurate dementia identification from electronic health records. medRxiv. https://doi.org/10.64898/2026.01.24.26344477
Vancouver
Wang L, Liu B, Yang R, Chuang Y, Estiri H, Murphy S, et al. A retrieval-augmented generation large language model framework for accurate dementia identification from electronic health records. medRxiv. 2026. doi:10.64898/2026.01.24.26344477.
BibTeX
@article{wang2026-789fc96e,
author = {Wang, L. and Liu, B. and Yang, R. and Chuang, Y. and Estiri, H. and Murphy, S. and Zhou, L. and Marshall, G. A.},
title = {{A retrieval-augmented generation large language model framework for accurate dementia identification from electronic health records}},
journal = {medRxiv},
year = {2026},
doi = {10.64898/2026.01.24.26344477},
url = {https://doi.org/10.64898/2026.01.24.26344477}
}