Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

A deep joint-learning proteomics model for diagnosis of six conditions associated with dementia

An, Lijun LU orcid ; Pichet Binette, Alexa LU ; Hristovska, Ines LU ; Vilkaite, Gabriele LU ; Xiao, Yu LU ; Zendehdel, Romina LU ; Dong, Zijian ; Smets, Bart ; Saloner, Rowan and Tasaki, Shinya , et al. (2026) In Nature Medicine
Abstract

Co-pathology is a common feature of neurodegenerative diseases that complicates diagnosis, treatment and clinical management. However, sensitive, specific and scalable biomarkers for in vivo pathological diagnosis are not available for most neurodegenerative neuropathologies. Here we present Proteomics-based Artificial Intelligence for Dementia Diagnosis (ProtAIDe-Dx), a deep joint-learning model on 17,187 patients and controls (age of 70.3 ± 11.5 years, 53.2% female), that uses plasma proteomics to provide simultaneous probabilistic diagnosis across 6 conditions associated with dementia in aging. ProtAIDe-Dx achieves cross-validated balanced classification accuracy of 70–95% and area under the curve of >78% across all conditions.... (More)

Co-pathology is a common feature of neurodegenerative diseases that complicates diagnosis, treatment and clinical management. However, sensitive, specific and scalable biomarkers for in vivo pathological diagnosis are not available for most neurodegenerative neuropathologies. Here we present Proteomics-based Artificial Intelligence for Dementia Diagnosis (ProtAIDe-Dx), a deep joint-learning model on 17,187 patients and controls (age of 70.3 ± 11.5 years, 53.2% female), that uses plasma proteomics to provide simultaneous probabilistic diagnosis across 6 conditions associated with dementia in aging. ProtAIDe-Dx achieves cross-validated balanced classification accuracy of 70–95% and area under the curve of >78% across all conditions. The model’s diagnostic probabilities highlighted subgroups of patients with co-pathologies and were associated with pathology-specific biomarkers in an external memory clinic sample, even among individuals without cognitive impairment. Model interpretation revealed a suite of protein networks marking shared and specific biological processes across diseases and identified novel and previously described proteins discriminating each diagnosis. ProtAIDe-Dx significantly improved biomarker-based differential diagnosis in a memory clinic sample, pinpointing proteins leading to diagnostic decisions at an individual level. Together, this work highlights the promise of plasma proteomics to improve patient-level diagnostic workup with a single blood draw.

(Less)
Please use this url to cite or link to this publication:
author
; ; ; ; ; ; ; ; and , et al. (More)
; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; and (Less)
author collaboration
organization
publishing date
type
Contribution to journal
publication status
epub
subject
in
Nature Medicine
publisher
Nature Publishing Group
external identifiers
  • scopus:105035320475
  • pmid:41917159
ISSN
1078-8956
DOI
10.1038/s41591-026-04303-y
language
English
LU publication?
yes
id
5939807e-70cc-4012-9b41-c8e01cafad30
date added to LUP
2026-05-13 12:18:43
date last changed
2026-05-27 13:13:33
@article{5939807e-70cc-4012-9b41-c8e01cafad30,
  abstract     = {{<p>Co-pathology is a common feature of neurodegenerative diseases that complicates diagnosis, treatment and clinical management. However, sensitive, specific and scalable biomarkers for in vivo pathological diagnosis are not available for most neurodegenerative neuropathologies. Here we present Proteomics-based Artificial Intelligence for Dementia Diagnosis (ProtAIDe-Dx), a deep joint-learning model on 17,187 patients and controls (age of 70.3 ± 11.5 years, 53.2% female), that uses plasma proteomics to provide simultaneous probabilistic diagnosis across 6 conditions associated with dementia in aging. ProtAIDe-Dx achieves cross-validated balanced classification accuracy of 70–95% and area under the curve of &gt;78% across all conditions. The model’s diagnostic probabilities highlighted subgroups of patients with co-pathologies and were associated with pathology-specific biomarkers in an external memory clinic sample, even among individuals without cognitive impairment. Model interpretation revealed a suite of protein networks marking shared and specific biological processes across diseases and identified novel and previously described proteins discriminating each diagnosis. ProtAIDe-Dx significantly improved biomarker-based differential diagnosis in a memory clinic sample, pinpointing proteins leading to diagnostic decisions at an individual level. Together, this work highlights the promise of plasma proteomics to improve patient-level diagnostic workup with a single blood draw.</p>}},
  author       = {{An, Lijun and Pichet Binette, Alexa and Hristovska, Ines and Vilkaite, Gabriele and Xiao, Yu and Zendehdel, Romina and Dong, Zijian and Smets, Bart and Saloner, Rowan and Tasaki, Shinya and Xu, Ying and Krish, Varsha and Imam, Farhad and Janelidze, Shorena and van Westen, Danielle and Stomrud, Erik and Whelan, Christopher D. and Palmqvist, Sebastian and Ossenkoppele, Rik and Mattsson-Carlgren, Niklas and Hansson, Oskar and Vogel, Jacob W.}},
  issn         = {{1078-8956}},
  language     = {{eng}},
  publisher    = {{Nature Publishing Group}},
  series       = {{Nature Medicine}},
  title        = {{A deep joint-learning proteomics model for diagnosis of six conditions associated with dementia}},
  url          = {{http://dx.doi.org/10.1038/s41591-026-04303-y}},
  doi          = {{10.1038/s41591-026-04303-y}},
  year         = {{2026}},
}