Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Machine learning prediction of tau-PET in Alzheimer's disease using plasma, MRI, and clinical data

Karlsson, Linda LU orcid ; Vogel, Jacob LU ; Arvidsson, Ida LU orcid ; Åström, Kalle LU orcid ; Strandberg, Olof LU ; Seidlitz, Jakob ; Bethlehem, Richard A.I. ; Stomrud, Erik LU orcid ; Ossenkoppele, Rik LU and Ashton, Nicholas J. , et al. (2025) In Alzheimer's and Dementia 21(2).
Abstract

INTRODUCTION: Tau positron emission tomography (PET) is a reliable neuroimaging technique for assessing regional load of tau pathology in the brain, but its routine clinical use is limited by cost and accessibility barriers. METHODS: We thoroughly investigated the ability of various machine learning models to predict clinically useful tau-PET composites (load and laterality index) from low-cost and non-invasive features, for example, clinical variables, plasma biomarkers, and structural magnetic resonance imaging (MRI). RESULTS: Models including plasma biomarkers yielded the most accurate predictions of tau-PET burden (best model: R-squared = 0.66–0.69), with especially high contribution from plasma phosphorylated tau-217 (p-tau217).... (More)

INTRODUCTION: Tau positron emission tomography (PET) is a reliable neuroimaging technique for assessing regional load of tau pathology in the brain, but its routine clinical use is limited by cost and accessibility barriers. METHODS: We thoroughly investigated the ability of various machine learning models to predict clinically useful tau-PET composites (load and laterality index) from low-cost and non-invasive features, for example, clinical variables, plasma biomarkers, and structural magnetic resonance imaging (MRI). RESULTS: Models including plasma biomarkers yielded the most accurate predictions of tau-PET burden (best model: R-squared = 0.66–0.69), with especially high contribution from plasma phosphorylated tau-217 (p-tau217). MRI variables were the best predictors of asymmetric tau load between the two hemispheres (best model: R-squared = 0.28–0.42). The models showed high generalizability to external test cohorts with data collected at multiple sites. Through a proof-of-concept two-step classification workflow, we also demonstrated possible model translations to a clinical setting. DISCUSSION: This study highlights the promising and limiting aspects of using machine learning to predict tau-PET from scalable cost-effective variables, with findings relevant for clinical settings and future research. Highlights: Accessible variables showed potential in estimating tau tangle load and distribution. Plasma phosphorylated tau-217 (p-tau217) and magnetic resonance imaging (MRI) were the best predictors of different tau-PET (positron emission tomography) composites. Machine learning models demonstrated high generalizability across AD cohorts.

(Less)
Please use this url to cite or link to this publication:
author
; ; ; ; ; ; ; ; and , et al. (More)
; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; and (Less)
organization
publishing date
type
Contribution to journal
publication status
published
subject
in
Alzheimer's and Dementia
volume
21
issue
2
article number
e14600
publisher
Wiley
external identifiers
  • scopus:85219624779
  • pmid:39985487
ISSN
1552-5260
DOI
10.1002/alz.14600
language
English
LU publication?
yes
additional info
Publisher Copyright: © 2025 The Author(s). Alzheimer's & Dementia published by Wiley Periodicals LLC on behalf of Alzheimer's Association.
id
02b0084a-5b2c-4195-a9c7-50c07b7adc1b
date added to LUP
2025-05-14 10:40:24
date last changed
2025-07-09 16:42:16
@article{02b0084a-5b2c-4195-a9c7-50c07b7adc1b,
  abstract     = {{<p>INTRODUCTION: Tau positron emission tomography (PET) is a reliable neuroimaging technique for assessing regional load of tau pathology in the brain, but its routine clinical use is limited by cost and accessibility barriers. METHODS: We thoroughly investigated the ability of various machine learning models to predict clinically useful tau-PET composites (load and laterality index) from low-cost and non-invasive features, for example, clinical variables, plasma biomarkers, and structural magnetic resonance imaging (MRI). RESULTS: Models including plasma biomarkers yielded the most accurate predictions of tau-PET burden (best model: R-squared = 0.66–0.69), with especially high contribution from plasma phosphorylated tau-217 (p-tau217). MRI variables were the best predictors of asymmetric tau load between the two hemispheres (best model: R-squared = 0.28–0.42). The models showed high generalizability to external test cohorts with data collected at multiple sites. Through a proof-of-concept two-step classification workflow, we also demonstrated possible model translations to a clinical setting. DISCUSSION: This study highlights the promising and limiting aspects of using machine learning to predict tau-PET from scalable cost-effective variables, with findings relevant for clinical settings and future research. Highlights: Accessible variables showed potential in estimating tau tangle load and distribution. Plasma phosphorylated tau-217 (p-tau217) and magnetic resonance imaging (MRI) were the best predictors of different tau-PET (positron emission tomography) composites. Machine learning models demonstrated high generalizability across AD cohorts.</p>}},
  author       = {{Karlsson, Linda and Vogel, Jacob and Arvidsson, Ida and Åström, Kalle and Strandberg, Olof and Seidlitz, Jakob and Bethlehem, Richard A.I. and Stomrud, Erik and Ossenkoppele, Rik and Ashton, Nicholas J. and Zetterberg, Henrik and Blennow, Kaj and Palmqvist, Sebastian and Smith, Ruben and Janelidze, Shorena and La Joie, Renaud and Rabinovici, Gil D. and Binette, Alexa Pichet and Mattsson-Carlgren, Niklas and Hansson, Oskar}},
  issn         = {{1552-5260}},
  language     = {{eng}},
  number       = {{2}},
  publisher    = {{Wiley}},
  series       = {{Alzheimer's and Dementia}},
  title        = {{Machine learning prediction of tau-PET in Alzheimer's disease using plasma, MRI, and clinical data}},
  url          = {{http://dx.doi.org/10.1002/alz.14600}},
  doi          = {{10.1002/alz.14600}},
  volume       = {{21}},
  year         = {{2025}},
}