Deep Learning for AD Risk Screening and Progression Prediction
14 affiliations
- 1State Key Laboratory of Cognitive Science and Mental Health Institute of Psychology Chinese Academy of Sciences Beijing People's Republic of China
- 2Department of Psychology University of the Chinese Academy of Sciences Beijing People's Republic of China
- 3Donders Centre for Cognitive Neuroimaging Donders Institute for Brain Cognition and Behaviour Radboud University Nijmegen the Netherlands
- 4Department of Psychology and Neuroscience University of North Carolina At Chapel Hill Chapel Hill North Carolina USA
- 5Department of Neurology National Clinical Research Center for Geriatric Diseases XuanWu Hospital of Capital Medical University Beijing People's Republic of China
- 6Department of Radiology and Nuclear Medicine Xuanwu Hospital Capital Medical University Beijing People's Republic of China
- 7Department of Neurology Beijing Chaoyang Hospital Capital Medical University Beijing People's Republic of China
- 8Department of Neurology of the Affiliated Hospital of Hangzhou Normal University Hangzhou People's Republic of China
- 9Department of Geriatrics of Changji People's Hospital Changji People's Republic of China
- 10Department of Child and Adolescent Psychiatry New York University Grossman School of Medicine New York New York USA
- 11Imaging Genetics Center Mark & Mary Stevens Institute for Neuroimaging & Informatics Keck School of Medicine University of Southern California Los Angeles California USA
- 12School of Biomedical Engineering Hainan University Haikou People's Republic of China
- 13Department of Psychological and Cognitive Sciences Tsinghua University Beijing People's Republic of China
- 14Research Center for Culture and Psychology Tsinghua University Beijing People's Republic of China
Abstract
The rise of Alzheimer's disease (AD) immunotherapies underscores the need for accessible, time‐efficient biomarkers for early diagnosis. MRI‐based deep learning shows promise in this regard, yet few models have undergone comprehensive clinical validation. Using the SILCODE cohort (722 Chinese participants, 1,105 scans), we test the transportability of our classifier (pretrained on Western samples) to an independent Chinese cohort without retraining. The associations between risk scores and disease progression, cognitive performance, and plasma biomarkers are examined. In addition, an interpretable framework is developed to generate individualized brain risk maps and identify subtypes. The model classifies AD with an AUC of 0.913 and 95.2% sensitivity, identifying 90.5% of high‐risk individuals 2–3 years before clinical diagnosis. Risk scores correlate significantly with cognitive and plasma indices, particularly p‐tau 217. Brain risk maps reveal AD subtypes, including a subgroup with rapid cognitive decline. These findings establish a population‐robust, clinically relevant MRI‐based model for early AD detection and risk stratification.
Related
- Time-varying risk prediction for oral potentially malignant disorders using deep learning
- Traditional Machine Learning versus Deep Learning Approaches for Multi-Outcome Perioperative Risk Prediction
- Physically aligned forest fire risk prediction: A deep learning framework coupling fuel and climate multivariate factors
- Machine learning and deep learning–based prediction of hypertension and analysis of its major risk factors in Bangladesh
- Longitudinal Imaging and Temporal Deep Learning for Dynamic Risk Prediction in Lung Cancer: Current Progress and Future Perspectives.
BibTeX
@article{lu2026deep,
title = {{Deep Learning for AD Risk Screening and Progression Prediction}},
author = {Bin Lu and Yan‐Rong Chen and Rui‐Xian Li and Ming‐Kai Zhang and Shao‐Zhen Yan and Guan‐Qun Chen and Jing‐Hua Wang and Wen‐Hui Chai and Francisco Xavier Castellanos and Paul M. Thompson and Jie Lu and Ying Han and Chao‐Gan Yan},
journal = {Advanced Science},
year = {2026},
month = nov,
doi = {10.1002/advs.78537},
url = {https://doi.org/10.1002/advs.78537},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Deep Learning for AD Risk Screening and Progression Prediction
AU - Bin Lu
AU - Yan‐Rong Chen
AU - Rui‐Xian Li
AU - Ming‐Kai Zhang
AU - Shao‐Zhen Yan
AU - Guan‐Qun Chen
AU - Jing‐Hua Wang
AU - Wen‐Hui Chai
AU - Francisco Xavier Castellanos
AU - Paul M. Thompson
AU - Jie Lu
AU - Ying Han
AU - Chao‐Gan Yan
PY - 2026
DA - 2026/11/05
JO - Advanced Science
DO - 10.1002/advs.78537
UR - https://doi.org/10.1002/advs.78537
AB - The rise of Alzheimer's disease (AD) immunotherapies underscores the need for accessible, time‐efficient biomarkers for early diagnosis. MRI‐based deep learning shows promise in this regard, yet few models have undergone comprehensive clinical validation. Using the SILCODE cohort (722 Chinese participants, 1,105 scans), we test the transportability of our classifier (pretrained on Western samples) to an independent Chinese cohort without retraining. The associations between risk scores and disease progression, cognitive performance, and plasma biomarkers are examined. In addition, an interpretable framework is developed to generate individualized brain risk maps and identify subtypes. The model classifies AD with an AUC of 0.913 and 95.2% sensitivity, identifying 90.5% of high‐risk individuals 2–3 years before clinical diagnosis. Risk scores correlate significantly with cognitive and plasma indices, particularly p‐tau 217. Brain risk maps reveal AD subtypes, including a subgroup with rapid cognitive decline. These findings establish a population‐robust, clinically relevant MRI‐based model for early AD detection and risk stratification.
SN - 2198-3844
ER -