Using conventional 7T systems with 32-channel receive arrays and body gradients, the effective resolution of single-shot EPI is encoding-limited by the readout speed relative to the T2*-decay rate. Given the typical size of a human head, and the constraints set by peripheral nerve stimulation, this limits the resolution of single-shot EPI to approximately 0.8mm. Segmented EPI approaches can overcome this limitation at the cost of longer TRs (Jin and Kim 2008; Menon and Goodyear 1999) (which, however, might limit some neuroscience applications whose task design requires short TRs). The correspondingly higher achievable spatial resolution can be used to provide better understanding of the underlying localization specificity and contrast-generating mechanisms. Here, we discuss such a segmented 3D-EPI approach (Stirnberg et al. 2016; 2017) to exemplify the laminar and columnar specificity and vessel sensitivity in conventional GE-BOLD and VASO. A matrix of 316 × 316 × 26 was used with 2 shots per kz-segment (TR = 3.9s).
