Machine Learning–Based Sleep Electroencephalographic Brain Age Index and Dementia Risk: An Individual - Summary - MDSpire

Machine Learning–Based Sleep Electroencephalographic Brain Age Index and Dementia Risk: An Individual

  • By

  • Haoqi Sun

  • Sasha Milton

  • Yi Fang

  • Hash Brown Taha

  • Shreya Shiju

  • Robert J. Thomas

  • Wolfgang Ganglberger

  • Matthew P. Pase

  • Timothy Hughes

  • Shaun Purcell

  • Susan Redline

  • Katie L. Stone

  • Kristine Yaffe

  • M. Brandon Westover

  • Yue Leng

  • March 19, 2026

  • 0 min

Share

Objective:

To compute the brain age index (BAI) from sleep EEG microstructures and examine its association with incident dementia in community-dwelling populations, providing a clearer understanding of BAI.

Key Findings:
  • An older sleep EEG-based BAI is associated with increased dementia risk, suggesting a need for early intervention.
  • The association between BAI and dementia risk varies by age and sex, indicating the importance of personalized assessments.
  • Key dementia risk factors influence the relationship between BAI and dementia, highlighting the complexity of dementia risk.
Interpretation:

The findings suggest that sleep EEG-derived BAI may serve as a valuable biomarker for assessing dementia risk, highlighting the importance of sleep microstructure in cognitive aging.

Limitations:
  • The study's findings may not be generalizable beyond the included cohorts, particularly due to demographic differences.
  • Variability in dementia ascertainment methods across cohorts may affect results, potentially leading to inconsistencies in the findings.
Conclusion:

The study supports the potential of using sleep EEG-derived BAI as a personalized approach for early detection of dementia risk in community-dwelling individuals, emphasizing its clinical relevance.

Original Source(s)

Machine Learning–Based Sleep Electroencephalographic Brain Age Index and Dementia Risk: An Individual

  • by Haoqi Sun, Sasha Milton, Yi Fang, Hash Brown Taha, Shreya Shiju, Robert J. Thomas, Wolfgang Ganglberger, Matthew P. Pase, Timothy Hughes, Shaun Purcell, Susan Redline, Katie L. Stone, Kristine Yaffe, M. Brandon Westover, Yue Leng

  • March 19, 2026

Related Content

The interplay of sleep architecture and exercise in executive function of middle-aged and older adults

The clinical value of multimodal neuroimaging in monoclonal antibody therapy for Alzheimer's disease

Could Digital Twins Forecast Forensic Risk?

Researchers outline a proposed three-level framework spanning patient monitoring, ward operations, and forensic system planning.