Maria-Eleni Dounavi, Audrey Low, Graciela Muniz Terrera, Karen Ritchie, Craig W. Ritchie, Li Su, Hugh S. Markus, John T O’Brien.

Brain Communications (2022)

DOI: https://academic.oup.com/braincomms/article/4/3/fcac116/6580683

Summary

Structural brain magnetic resonance imaging scans are typically used to measure properties of brain areas such as their volume or thickness. However, the intensity values of the acquired scans, can further be analysed and ‘texture’ can be quantified. Textural analysis examines variations in the image intensity and quantifies properties such as homogeneity or contrast of the image at a regional level. In this study we have measured textural properties of brain scans from PREVENT-Dementia participants and investigated if these were connected to performance in a reaction time task, to a marker of cerebrovascular pathology (white matter hyperintensity -WMH volume) and to a cardiovascular dementia risk score (CAIDE). Our findings suggest that textural measures might capture subtle underlying damage in normal appearing brain areas and could be more sensitive measures of early changes compared to WMH volume.