Persistent exposure to highly pulsatile blood can damage the brain’s microvasculature. A convenient method for measuring cerebral pulsatility would allow investigation into its relationship with vascular dysfunction and cognitive decline. In this work, we propose a convolutional neural network (CNN) based deep learning solution to estimate cerebral pulsatility using only the frequency content from BOLD MRI scans. Various frequency component inputs were assessed, and echo time dependence was evaluated with a 5-fold cross-validation. Pulsatility was estimated from BOLD MRI data acquired on a different scanner to assess generalizability. The CNN reliably estimated pulsatility and was robust to various scan parameters.

No abstract

No abstract

Anatomical T1-weighted imaging allows us to examine relationships of regional morphology across the brain through structural covariance. Here, we investigated structural covariance stability using decreasing amounts of T1-weighted imaging data from the UK Biobank. Starting from 1,753 individuals, we found that it is possible to drastically reduce the sample and still maintain adequate stability (78% agreement with ~87 individuals). We note, however, that stability was regionally variable; lateral and cortical regions were least affected by sample size while medial and subcortical regions were most affected. These findings may inform sample size considerations for MRI-based structural covariance in large population studies.

Measuring intracranial vascular compliance (VC) may help characterize a patient’s resilience to Alzheimer’s disease and dementia. In this study, we estimate VC from a BOLD MRI cerebrovascular reactivity (CVR) protocol by measuring cardiac-related BOLD pulsatility and end-tidal CO2 during periods of hypercapnia and normocapnia. We measured high VC in regions proximal to major cerebral arteries like the thalamus and insula, and low VC in the parietal lobe. This study demonstrates the feasibility of extracting additional information from BOLD-CVR experiments and suggests regional differences in VC. Future work will include a second cohort to estimate VC in different patient populations.