Abstract #4337

Processing cerebrovascular reactivity data using shift-invariant dictionary learning

Emilie Sleight^1,2, Michael S Stringer^1,2, Ian Marshall^1,2, Joanna M Wardlaw^1,2, Sotirios A Tsaftaris³, and Michael J Thrippleton^1,2

¹Centre for Clinical Brain Sciences, University of Edinburgh, Edinburgh, United Kingdom, ²UK Dementia Research Institute, Edinburgh, United Kingdom, ³Institute for Digital Communications, University of Edinburgh, Edinburgh, United Kingdom

The BOLD response to a hypercapnic challenge, i.e. cerebrovascular reactivity (CVR), may vary between individuals and tissue types. Linear regression (GLM) between the BOLD signal and the end-tidal CO₂ is the most common CVR processing method but does not allow for different haemodynamic responses across the brain. We propose to use shift-invariant dictionary learning (SIDL) as a promising method to enable data-driven extraction of BOLD response(s). We show that CVR and delay estimates from SIDL are comparable to estimates from GLM. Future work will focus on reducing the effect of drift on SIDL estimates.

This abstract and the presentation materials are available to members only; a login is required.

Join Here