Skip to main content
Fig. 1 | Fluids and Barriers of the CNS

Fig. 1

From: Filter exchange imaging with crusher gradient modelling detects increased blood–brain barrier water permeability in response to mild lung infection

Fig. 1

Imaging BBB water exchange. a FEXI pulse sequence diagram. The filter block consists of a 90° excitation pulse with associated slab-selection gradient (dark orange), followed by a fixed pair of diffusion-encoding gradients (with filter gradient amplitude (gf), duration (δf) and diffusion time (f), light orange) separated by a slab selective refocussing pulse. The mixing block, for encoding water exchange during varying mixing time tm, consists of crusher gradients (grey) (with crusher gradient amplitude (gc) dependent on slice thickness (∆z)), slice encoding gradients (dark green) associated with the RF pulses, and a spoiler gradient (black) to null unwanted transverse magnetization. The detection block, for signal readout, consists of variable diffusion encoding gradients (with gradient amplitude (g), duration (δ) and diffusion time (), light green) followed by an echo planar imaging (EPI) readout. Adapted from:[43]. b, c AXR model for water exchange across BBB. A two-compartment model of water-exchange can be parameterised by forward/ backwards exchange rates (kin / kout), intra/extra-vascular diffusivities (Di/ De) and intra/extra-vascular signal fractions (fi/ fe). Measurements of the ADC are first acquired at equilibrium (ADCeq) without the diffusion filter. The diffusion filter is then applied, which supresses the fast-diffusing intravascular water, leading to a reduction in the measured ADC (ADC'). As the mixing time increases, water-exchange between the two compartments results in a recovery of the ADC’ back to ADCeq. The crusher gradients (higher magnitude with smaller ∆z) introduce additional diffusion-weighting that results in an undesirable suppression of the ADC'(tm) recovery. The apparent exchange rate (AXR) models the rate of ADC' recovery as an exponential. Since this model does not account for diffusion weighting caused by the longitudinal storage crusher gradients, it underestimates the true exchange rate when these crusher gradients contribute significantly to the overall diffusion weighting of the filter block

Back to article page