Purpose To develop and optimise a 3D black-blood R2* mapping sequence for imaging the carotid artery wall, using optimal blood suppression and k-space view ordering.

Methods Two different blood suppression preparation methods were used; Delay Alternating with Nutation for Tailored Excitation (DANTE) and improved Motion Sensitive Driven Equilibrium (iMSDE) were each combined with a three-dimensional (3D) multi-echo Fast Spoiled GRadient echo (ME-FSPGR) readout. Three different k-space view-order designs: Radial Fan-beam Encoding Ordering (RFEO), Distance-Determined Encoding Ordering (DDEO) and Centric Phase Encoding Order (CPEO) were investigated. The sequences were evaluated through Bloch simulation and in a cohort of twenty volunteers. The vessel wall Signal-to-Noise Ratio (SNR), Contrast-to-Noise Ratio (CNR) and R2*, and the sternocleidomastoid muscle R2* were measured and compared. Different numbers of acquisitions-per-shot (APS) were evaluated to further optimise the effectiveness of blood suppression.

Results All sequences resulted in comparable R2* measurements to a conventional, i.e. non-blood suppressed sequence in the sternocleidomastoid muscle of the volunteers. Both Bloch simulations and volunteer data showed that DANTE has a higher signal intensity and results in a higher image SNR than iMSDE. Blood suppression efficiency was not significantly different when using different k-space view orders. Smaller APS achieved better blood suppression.

Conclusion The use of blood-suppression preparation methods does not affect the measurement of R2*. DANTE prepared ME-FSPGR sequence with a small number of acquisitions-per-shot can provide high quality black-blood R2* measurements of the carotid vessel wall.