Machine-learning-derived heart and brain age are independently associated with cognition
Olena Iakunchykova1, Henrik Schirmer2,3, Torgil Vangberg4,5
1Center for Lifespan Changes in Brain and Cognition, Department of Psychology, University of Oslo, Oslo, Norway.
European Journal of Neurology
|May 31, 2023
Summary
A new heart age biomarker derived from electrocardiograms is linked to cognitive function. This heart delta age (HDA) shows associations with cognitive tests, partly mediated by brain delta age (BDA).
Area of Science:
- Cardiology
- Neurology
- Biomarker Development
Background:
- A novel heart age biomarker has been developed utilizing deep neural networks applied to electrocardiograms.
- The study investigates the potential association between this heart age biomarker and cognitive function.
Purpose of the Study:
- To determine if a deep neural network-derived heart age biomarker is associated with cognitive function.
- To explore the relationship between heart delta age (HDA), brain delta age (BDA), and cognitive test performance.
Main Methods:
- Heart age was estimated from 12-lead electrocardiograms in a population-based sample (N=7779, ages 40-85).
- Associations between heart delta age (HDA) and cognitive test scores were analyzed, adjusting for cardiovascular risk factors.
- Mediation analysis examined the relationship between HDA, brain delta age (BDA), and cognitive test scores.
Main Results:
- Significant associations were found between HDA and scores on the Word test, Digit Symbol Coding Test, and tapping test.
- HDA was correlated with BDA (Pearson's r=0.12, p=0.0001).
- BDA mediated 13% of the effect of HDA on the tapping test score.
Conclusions:
- Heart delta age, reflecting cumulative life-long exposures, is associated with brain age.
- HDA is associated with cognitive function, with a minimal portion of this effect being explained by BDA.
