Transfer learning for cognitive reserve quantification

Topic: Brain imaging, connectivity and assessment

Cognitive reserve (CR) has been introduced to explain individual differences in susceptibility to cognitive or functional impairment in the presence of age or pathology. We developed a deep learning model to quantify the CR as residual variance in memory performance using the Structural Magnetic Resonance Imaging (sMRI) data from a lifespan healthy cohort. The generalizability of the sMRI-based deep learning model was tested in two independent healthy and Alzheimer's cohorts using transfer learning framework. Structural MRIs were collected from three cohorts: 495 healthy adults (age: 20-80) from RANN, 620 healthy adults (age: 36-100) from lifespan Human Connectome Project Aging (HCPA), and 941 adults (age: 55-92) from Alzheimer's Disease Neuroimaging Initiative (ADNI). Region of interest (ROI)-specific cortical thickness and volume measures were extracted using the Desikan-Killiany Atlas. CR was quantified by residuals which subtract the predicted memory from the true memory. Cascade neural network (CNN) models were used to train RANN dataset for memory prediction. Transfer learning was applied to transfer the T1 imaging-based model from source domain (RANN) to the target domains (HCPA or ADNI). The CNN model trained on the RANN dataset exhibited strong linear correlation between true and predicted memory based on the T1 cortical thickness and volume predictors. In addition, the model generated from healthy lifespan data (RANN) was able to generalize to an independent healthy lifespan data (HCPA) and older demented participants (ADNI) across different scanner types. The estimated CR was correlated with CR proxies such education and IQ across all three datasets. The current findings suggest that the transfer learning approach is an effective way to generalize the residual-based CR estimation. It is applicable to various diseases and may flexibly incorporate different imaging modalities such as fMRI and PET, making it a promising tool for scientific and clinical purposes.

Key findings

• The current findings suggest that the transfer learning approach is an effective way to generalize the residual-based CR estimation.

New to this subject? Start with our plain-English guide: Risk factors, ranked →

Study details

• Participants941
• Evidence strengthModerate

Read the original

• DOI: 10.1016/j.neuroimage.2022.119353
• PubMed: 35667639
• Original source →

Free copy unknown: no free-copy check has an answer for this record yet. That does not mean it is paywalled.

@article{zhu2022-35667639,
  author  = {Zhu, Xi and Liu, Yi and Habeck, Christian G and Stern, Yaakov and Lee, Seonjoo and {For-The-Alzheimer's-Disease-Neuroimaging-Initiative}},
  title   = {{Transfer learning for cognitive reserve quantification}},
  journal = {NeuroImage},
  year    = {2022},
  doi     = {10.1016/j.neuroimage.2022.119353},
  pmid    = {35667639},
  url     = {https://doi.org/10.1016/j.neuroimage.2022.119353}
}

Related studies in Risk factors

• Understanding Social Determinants of Health and Cognitive Impairment Risk in Diverse Populations Using Machine Learning: A HABS‐HD Study Alzheimer s & Dementia · 2025-12-01 · usefulness score 38/100
• Predicting low premorbid cognitive ability with social determinants: A machine learning approach Journal of Aging Research and Lifestyle · 2026-01-01 · usefulness score 31/100
• Integrating Machine Learning and Environmental and Genetic Risk Factors for the Early Detection of Preclinical Alzheimer's Disease The journals of gerontology. Series B, Psychological sciences and social sciences · 2025-02-19 · usefulness score 65/100
• Insulin and the insulin receptor in experimental models of learning and memory European journal of pharmacology · 2004-04-01 · usefulness score 48/100
• Alzheimer’s Disease Dementia Prevalence in the United States: A County-Level Spatial Machine Learning Analysis American Journal of Alzheimer s Disease & Other Dementias® · 2025-04-21 · usefulness score 42/100
• Mayo Normative Studies: regression‐based normative data for remote self‐administration of the Stricker Learning Span, Symbols Test and Mayo Test Drive Screening Battery Composite for ages 33‐100 and validation in MCI and dementia Alzheimer s & Dementia · 2024-12-01 · usefulness score 42/100

Similar studies

• Deep learning-based brain age prediction in normal aging and dementia 2022 · Digital health tools
• MRI signatures of brain age and disease over the lifespan based on a deep brain network and 14 468 individuals worldwide 2020 · Digital health tools
• A neuroimaging approach to capture cognitive reserve: Application to Alzheimer's disease 2017 · Risk factors
• Classification and Visualization of Alzheimer’s Disease using Volumetric Convolutional Neural Network and Transfer Learning 2019 · Brain scans and imaging
• The Brain Chart of Aging: Machine-learning analytics reveals links between brain aging, white matter disease, amyloid burden, and cognition in the iSTAGING consortium of 10,216 harmonized MR scans 2021 · Digital health tools
• Convolutional Neural Networks-Based MRI Image Analysis for the Alzheimer's Disease Prediction From Mild Cognitive Impairment 2018 · Digital health tools

More on Risk factors

• A review that covers this ground Cognitive reserve in individuals with frontotemporal dementia: a systematic review Journal of clinical and experimental neuropsychology · 2024-10-17 · usefulness score 82/100
• Newest in this subject Neuroscience-Informed Psychoeducation for Dementia: From Network Degeneration to Compassionate and Skilled Care PsyArXiv · 2026-10-28 · Preprint — not yet peer-reviewed · usefulness score 16/100
• Newest in this subject Geographic Heterogeneity in the Association Between Schooling and Cognitive Change Among Older Adults in India medRxiv · 2026-10-28 · Preprint — not yet peer-reviewed · usefulness score 16/100

See all studies in Risk factors →