A model of brain function to predict tau across Alzheimer’s disease and related disorders
Abstract
We developed a machine learning framework grounded in a computational model of brain function and FDG-PET imaging to predict Alzheimer’s disease (AD)-type tau across AD and related disorders (ADRD). The model was trained on 1088 research participants spanning normal aging and ADRD with FDG-PET and flortaucipir-PET (FTP-PET). Validation was conducted in: (1) a longitudinal subset of the training cohort ( n = 215), (2) Alzheimer’s Disease Neuroimaging Initiative ( n = 295), and (3) a heretogeneous clinical cohort ( n = 264; n = 28 with FTP-PET). Results showed that the model accurately predicted FTP-PET when applied at the participant-level ( R 2 ~ 0.34-0.71). It performed well in clinical scenarios, including classification of low versus high tau stages (79–95% accuracy) and prediction of tau in a heterogeneous clinical cohort without FTP-PET. These findings advance the understanding of the relationship between neurodegeneration and AD-type tau and provide a framework for estimating AD-type tau across ADRD.
Related
- An effective and deep learning model to predict Alzheimer’s disease at an early stage from audio data
- Item‐level verbal fluency psycholinguistic metrics predict preclinical Alzheimer's disease
- Cognitive function and return to work: protocol for a longitudinal study investigating cognitive endurance in patients with stress-related disorders in primary health care in Sweden
- Olfactory Function in Preclinical Alzheimer's Disease: A Multi-Cohort Study
- Potential links between PCOS/PMOS and Alzheimer disease and related dementias
BibTeX
@article{corriveaulecavalier2026model,
title = {{A model of brain function to predict tau across Alzheimer’s disease and related disorders}},
author = {Nick Corriveau-Lecavalier and Gemeng Zhang and Hugo Botha and Leland R. Barnard and Heather J. Wiste and Ellen Dicks and Venkatsampath R. Gogineni and Jeyeon Lee and Irene Sintini and Matthew L. Senjem and Christopher G. Schwarz and Petrice M. Cogswell and Derek R. Johnson and Jeffrey L. Gunter and Michael G. Kamykowski and Jonathan Graff-Radford and Vijay K. Ramanan and Julie A. Fields and Mary M. Machulda and Nikki H. Stricker and Bradley F. Boeve and Val J. Lowe and Kejal Kantarci and David S. Knopman and Jennifer L. Whitwell and Keith A. Josephs and Ronald C. Petersen and Clifford R. Jack and David T. Jones and or the Alzheimer’s Disease Neuroimaging Initiative},
journal = {Nature Communications},
year = {2026},
month = oct,
doi = {10.1038/s41467-026-78300-7},
url = {https://www.nature.com/articles/s41467-026-78300-7},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - A model of brain function to predict tau across Alzheimer’s disease and related disorders
AU - Nick Corriveau-Lecavalier
AU - Gemeng Zhang
AU - Hugo Botha
AU - Leland R. Barnard
AU - Heather J. Wiste
AU - Ellen Dicks
AU - Venkatsampath R. Gogineni
AU - Jeyeon Lee
AU - Irene Sintini
AU - Matthew L. Senjem
AU - Christopher G. Schwarz
AU - Petrice M. Cogswell
AU - Derek R. Johnson
AU - Jeffrey L. Gunter
AU - Michael G. Kamykowski
AU - Jonathan Graff-Radford
AU - Vijay K. Ramanan
AU - Julie A. Fields
AU - Mary M. Machulda
AU - Nikki H. Stricker
AU - Bradley F. Boeve
AU - Val J. Lowe
AU - Kejal Kantarci
AU - David S. Knopman
AU - Jennifer L. Whitwell
AU - Keith A. Josephs
AU - Ronald C. Petersen
AU - Clifford R. Jack
AU - David T. Jones
AU - or the Alzheimer’s Disease Neuroimaging Initiative
PY - 2026
DA - 2026/10/02
JO - Nature Communications
DO - 10.1038/s41467-026-78300-7
UR - https://www.nature.com/articles/s41467-026-78300-7
AB - We developed a machine learning framework grounded in a computational model of brain function and FDG-PET imaging to predict Alzheimer’s disease (AD)-type tau across AD and related disorders (ADRD). The model was trained on 1088 research participants spanning normal aging and ADRD with FDG-PET and flortaucipir-PET (FTP-PET). Validation was conducted in: (1) a longitudinal subset of the training cohort ( n = 215), (2) Alzheimer’s Disease Neuroimaging Initiative ( n = 295), and (3) a heretogeneous clinical cohort ( n = 264; n = 28 with FTP-PET). Results showed that the model accurately predicted FTP-PET when applied at the participant-level ( R 2 ~ 0.34-0.71). It performed well in clinical scenarios, including classification of low versus high tau stages (79–95% accuracy) and prediction of tau in a heterogeneous clinical cohort without FTP-PET. These findings advance the understanding of the relationship between neurodegeneration and AD-type tau and provide a framework for estimating AD-type tau across ADRD.
SN - 2041-1723
ER -