Cerebral Perfusion and Cognitive Decline in Type 2 Diabetes

This entry covers the brain MRI cohort of a single-center, cross-sectional study at Beth Israel Deaconess Medical Center in Boston (study GE-75). The study asked how type 2 diabetes changes the regulation and regional distribution of cerebral blood flow in older adults, and whether poor glucose control adds to cognitive and motor decline. The MRI images are not part of the public release. PhysioNet hosts, under CC BY 4.0, a per-subject summary table with derived MRI measures, clinical, laboratory, eye and cognitive data, plus transcranial Doppler, ECG and foot pressure recordings.

Composition

The project page describes 70 adults with type 2 diabetes treated for more than a year and 70 healthy controls, aged 50 to 85 and recruited from the local community. The released summary table has 88 subject rows: 75 labelled diabetic and 13 control, 50 men and 38 women, aged 50 to 83. The sex and age rows in the statistics are counted from that table, so they describe those 88 people and not the full cohort.

Acquisition

Participants spent two days at the hospital's clinical research center. On the second day, after Doppler testing of cerebral vasoreactivity to CO2 and posture changes, they had MRI at 3 Tesla on a GE scanner. The project page names T1- and T2-weighted imaging, continuous arterial spin labeling perfusion at rest and during CO2 challenges, and diffusion tensor imaging. The paper that PhysioNet asks users to cite also reports FLAIR. The study protocol additionally lists dual-echo fast spin echo and time-of-flight angiography as planned sequences.

Annotations

The summary table holds derived values only: regional gray and white matter volumes in a labelled template, white matter hyperintensity volumes per lobe and region, and regional perfusion measures under different conditions.

Known limitations

• MRI images, segmentations and perfusion maps are not shared, and no terms for them are published.
• The summary table covers 88 participants with a different diabetic to control split than the 70 and 70 on the project page; the release does not explain the gap.
• The cited paper (Novak et al. 2011) reports 147 participants and does not say how they map onto this release.