Related Experiment Video
Updated: Sep 13, 2025
Abbiategrasso Brain Bank Protocol for Collecting, Processing and Characterizing Aging Brains
Published on: June 3, 2020
Development and Validation of a Brain Aging Biomarker in Middle-Aged and Older Adults: Deep Learning Approach
Zihan Li1, Jun Li2, Jiahui Li1
1Center for Clinical Big Data and Analytics, The Second Affiliated Hospital and Department of Big Data in Health Science, School of Public Health, Zhejiang University School of Medicine, Zhejiang University, Hangzhou, China.
This study introduces a novel deep learning model, Brain Vision Graph Neural Network (BVGN), for precise brain aging assessment. The BVGN model accurately predicts cognitive decline and aids in early detection of neurodegenerative diseases like mild cognitive impairment (MCI).
Area of Science:
- Neuroimaging and Artificial Intelligence
- Biomarker Discovery for Neurodegeneration
- Computational Neuroscience
Background:
- Accurate brain aging assessment is vital for early detection of neurodegenerative disorders.
- Current MRI-based methods need improvement in capturing local brain morphology and topological structures.
Purpose of the Study:
- To develop and validate a deep learning framework (BVGN) for precise brain aging estimation.
- Incorporate connectivity and complexity for enhanced accuracy in aging prediction.
- Facilitate early identification of neurodegenerative diseases.
Main Methods:
- Utilized 5889 T1-weighted MRI scans from the Alzheimer's Disease Neuroimaging Initiative.
- Developed a novel Brain Vision Graph Neural Network (BVGN) with neurobiologically informed features.
- Validated model performance and generalization on an external UK Biobank dataset (N=34,352).
Main Results:
- BVGN achieved a mean absolute error (MAE) of 2.39 years, outperforming existing methods.
- The brain age gap showed significant differences across cognitive states (CN, MCI, Alzheimer's disease; P<.001).
- BVGN demonstrated superior discriminative capacity for MCI detection (AUC=0.885) compared to conventional markers.
Conclusions:
- BVGN provides a precise and generalizable framework for brain aging assessment.
- The derived brain age gap serves as a sensitive biomarker for early MCI identification and cognitive decline prediction.
- BVGN holds significant potential for clinical applications in neurodegenerative disease management.
