CARE-AD: a multi-agent large language model framework for Alzheimer’s disease prediction using longitudinal clinical notes
Abstract Large language models (LLMs) have shown promising capabilities across diverse domains, yet their application to complex clinical prediction tasks remains limited. In this study, we present CARE-AD (Collaborative Analysis and Risk Evaluation for Alzheimer’s Disease), a multi-agent LLM-based...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-08-01
|
| Series: | npj Digital Medicine |
| Online Access: | https://doi.org/10.1038/s41746-025-01940-4 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849225930075013120 |
|---|---|
| author | Rumeng Li Xun Wang Dan Berlowitz Jesse Mez Honghuang Lin Hong Yu |
| author_facet | Rumeng Li Xun Wang Dan Berlowitz Jesse Mez Honghuang Lin Hong Yu |
| author_sort | Rumeng Li |
| collection | DOAJ |
| description | Abstract Large language models (LLMs) have shown promising capabilities across diverse domains, yet their application to complex clinical prediction tasks remains limited. In this study, we present CARE-AD (Collaborative Analysis and Risk Evaluation for Alzheimer’s Disease), a multi-agent LLM-based framework for forecasting Alzheimer’s disease (AD) onset by analyzing longitudinal electronic health record (EHR) notes. CARE-AD assigns specialized LLM agents to extract signs and symptoms relevant to AD and conduct domain-specific evaluations—emulating a collaborative diagnostic process. In a retrospective evaluation, CARE-AD achieved higher accuracy (0.53 vs. 0.26–0.45) than baseline single-model approaches in predicting AD risk 10 years prior to the first recorded diagnosis code. These findings highlight the feasibility of using multi-agent LLM systems to support early risk assessment for AD and motivate further research on their integration into clinical decision support workflows. |
| format | Article |
| id | doaj-art-a51166a84b0d4d7f8373f652d745d414 |
| institution | Kabale University |
| issn | 2398-6352 |
| language | English |
| publishDate | 2025-08-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | npj Digital Medicine |
| spelling | doaj-art-a51166a84b0d4d7f8373f652d745d4142025-08-24T11:52:10ZengNature Portfolionpj Digital Medicine2398-63522025-08-01811910.1038/s41746-025-01940-4CARE-AD: a multi-agent large language model framework for Alzheimer’s disease prediction using longitudinal clinical notesRumeng Li0Xun Wang1Dan Berlowitz2Jesse Mez3Honghuang Lin4Hong Yu5Manning College of Information & Computer Sciences, University of Massachusetts AmherstMicrosoft CorporationCenter for Health Organization & Implementation Research, VA Bedford Health Care SystemChobanian & Avedisian School of Medicine, Boston UniversityDepartment of Medicine, UMass Chan Medical SchoolManning College of Information & Computer Sciences, University of Massachusetts AmherstAbstract Large language models (LLMs) have shown promising capabilities across diverse domains, yet their application to complex clinical prediction tasks remains limited. In this study, we present CARE-AD (Collaborative Analysis and Risk Evaluation for Alzheimer’s Disease), a multi-agent LLM-based framework for forecasting Alzheimer’s disease (AD) onset by analyzing longitudinal electronic health record (EHR) notes. CARE-AD assigns specialized LLM agents to extract signs and symptoms relevant to AD and conduct domain-specific evaluations—emulating a collaborative diagnostic process. In a retrospective evaluation, CARE-AD achieved higher accuracy (0.53 vs. 0.26–0.45) than baseline single-model approaches in predicting AD risk 10 years prior to the first recorded diagnosis code. These findings highlight the feasibility of using multi-agent LLM systems to support early risk assessment for AD and motivate further research on their integration into clinical decision support workflows.https://doi.org/10.1038/s41746-025-01940-4 |
| spellingShingle | Rumeng Li Xun Wang Dan Berlowitz Jesse Mez Honghuang Lin Hong Yu CARE-AD: a multi-agent large language model framework for Alzheimer’s disease prediction using longitudinal clinical notes npj Digital Medicine |
| title | CARE-AD: a multi-agent large language model framework for Alzheimer’s disease prediction using longitudinal clinical notes |
| title_full | CARE-AD: a multi-agent large language model framework for Alzheimer’s disease prediction using longitudinal clinical notes |
| title_fullStr | CARE-AD: a multi-agent large language model framework for Alzheimer’s disease prediction using longitudinal clinical notes |
| title_full_unstemmed | CARE-AD: a multi-agent large language model framework for Alzheimer’s disease prediction using longitudinal clinical notes |
| title_short | CARE-AD: a multi-agent large language model framework for Alzheimer’s disease prediction using longitudinal clinical notes |
| title_sort | care ad a multi agent large language model framework for alzheimer s disease prediction using longitudinal clinical notes |
| url | https://doi.org/10.1038/s41746-025-01940-4 |
| work_keys_str_mv | AT rumengli careadamultiagentlargelanguagemodelframeworkforalzheimersdiseasepredictionusinglongitudinalclinicalnotes AT xunwang careadamultiagentlargelanguagemodelframeworkforalzheimersdiseasepredictionusinglongitudinalclinicalnotes AT danberlowitz careadamultiagentlargelanguagemodelframeworkforalzheimersdiseasepredictionusinglongitudinalclinicalnotes AT jessemez careadamultiagentlargelanguagemodelframeworkforalzheimersdiseasepredictionusinglongitudinalclinicalnotes AT honghuanglin careadamultiagentlargelanguagemodelframeworkforalzheimersdiseasepredictionusinglongitudinalclinicalnotes AT hongyu careadamultiagentlargelanguagemodelframeworkforalzheimersdiseasepredictionusinglongitudinalclinicalnotes |