Multi-Cohort, Multi-Sequence Harmonisation For Cerebrovascular Brain Age
Imagine trying to compare apples and oranges when studying how our brains age. That’s often the challenge researchers face when combining brain scans from different studies, as variations between MRI scanners can make direct comparisons difficult. This challenge is particularly important when trying to estimate a “brain age,” which is the predicted age of a person’s brain based on imaging data. If your brain age is higher than your actual age, it suggests an accelerated aging process and might be linked to cognitive decline.
This study tackled the problem of combining diverse brain imaging data, especially focusing on a special type of MRI called Arterial Spin Labeling (ASL). ASL measures blood flow in the brain, giving us insights into “cerebrovascular health” – the health of the blood vessels in the brain. Since cardiovascular health plays a big role in how our brains age, including ASL data in brain age predictions is crucial. However, differences in how ASL scans are acquired across different research centers can introduce inconsistencies.
To overcome these inconsistencies, the researchers investigated several “harmonisation” methods. Think of harmonisation as a sophisticated way to standardize data collected from various sources, making them comparable. By applying these methods, the study significantly improved the reliability and accuracy of brain age predictions that incorporate blood flow information. This means that researchers can now more confidently combine data from multiple studies to get a clearer, more comprehensive understanding of how our blood vessels contribute to brain aging and its impact on thinking abilities.
Source: link to paper