Brain Aging in Patients With Cardiovascular Disease From the UK Biobank

Area of Science:

• Neuroimaging
• Cardiovascular Health
• Artificial Intelligence in Medicine

Background:

• Brain aging involves normal structural changes, but deviations can signal disease risk.
• The brain age gap (BAG) quantifies deviations from normal brain aging using machine learning.
• Previous studies show BAG's promise for neurological diseases and cardiovascular risk factors.

Purpose of the Study:

• To investigate the utility of BAG in quantifying brain changes associated with diagnosed cardiovascular diseases.
• To compare BAG in patients with cardiovascular diseases against healthy controls and neurological disease patients.
• To explore if brain regions used for BAG prediction differ across disease groups.

Main Methods:

• T1-weighted MRI scans from UK Biobank participants were used.
• A convolutional neural network (CNN) model was trained to predict biological brain age.
• BAG was calculated and compared across cardiovascular disease groups, healthy controls, and neurological disease patients.
• Saliency maps analyzed brain regions critical for CNN's age prediction.

Main Results:

• Significant differences in BAG distributions were found for 10 of 42 sex-specific cardiovascular disease groups compared to healthy participants.
• No significant differences were observed in the brain regions identified by saliency maps across groups.
• The CNN model predominantly relied on healthy brain aging patterns, even with cardiovascular diseases present.

Conclusions:

• The BAG serves as a sensitive imaging biomarker for detecting brain aging differences linked to specific cardiovascular diseases.
• Findings support the heart-brain axis theory, highlighting cardiovascular diseases' association with atypical brain aging.
• BAG analysis offers insights into the neurological impact of cardiovascular conditions.