Looping Star

FIGURE S1 Schematic illustration of Looping Star assuming NSpkPerLoop = 6 and NLoop = 3 (i.e. 1 FID and 2 equidistantechoes). One single segment of the pulse sequence is illustrated in Fig. 1A. RF excitation block pulses (bold red) are applied during the initial FID loop and turned off afterwards. The readout gradient has constant magnitude (bold black) but it’s direction increments by an in-plane rotation of 2π/NSpkPerLoop = π/4 from one spoke to the next. The corresponding k-space encoding is shown in Fig. 1B. Each excitation starts with a straight radial spoke (bold black) which subsequently accumulates to a regular hexagon (black). The selfrefocusing hexagon trajectories of the first two excitations are illustrated in blue and green, respectively. Figure 1C illustrates the magnitude k-space dephasing and refocusing over time. Each excitation generates one coherence (during the initial FID loop) which subsequently refocus (during the later gradient echo loops). In a way, Looping Star can be understood as time-multiplexed, 3D radial gradient echo imaging. For 3D spatial encoding, the planar k-space trajectory needs to be rotated as illustrated in Fig. 1D

FIGURE S2 Looping Star provides intrinsic coil sensitivity calibration. The single-channel PD-weighted FID images (top) are ideally suited for coil sensitivity calibration (bottom) as required for parallel imaging scan acceleration