Breath-held single shot turbo spin echo (SShTSE) is a widely used T2-weighted acquisition in abdominal imaging due to its speed combined with robustness to field inhomogeneities and motion^[1,2]. SShTSE is often performed with fat suppression to increase the conspicuity of the anatomical details, increasing the total scan times. Additionally, chemically selective saturation pulses used for fat suppression are sensitive to B0 inhomogeneity providing incomplete fat suppression^[3]. Alternatively, Dixon based techniques can be used to achieve uniform fat suppression, but at the expense of prolonged acquisition times^[4].

The purpose of this work is to implement a dual-echo SShTSE acquisition where the in-phase (IP) and out-of-phase (OP) echoes required for Dixon processing are acquired in the same repetition times, providing a true single shot acquisition with robust fat/water separation. In addition, the sequence is evaluated and compared against SShTSE with and without SPAIR on normal volunteers and patients.

SShTSE typically uses partial phase encoding to acquire the entire k-space in a single shot without significant T2 decay and utilizes Homodyne reconstruction to generate the final image. SShTSE can be modified to acquire both IP and OP echoes in the same repetition between each pair of refocusing pulses (fig. 1b). However, it enforces to use partial echo along the readout direction (fig. 1c) to acquire IP and OP echoes at optimal δt (e.g. 1.1 ms at 3T) for Dixon processing. This provides a unique challenge of acquiring the k-space with both partial echo and partial phase encoding for dual-echo Dixon SShTSE. In such scenario, the pulse sequence and the corresponding reconstruction were modified to perform zero filling along the frequency encoding direction and homodyne reconstruction along the phase encoding direction. Dual-echo Dixon SShTSE was implemented on a 3 T Ingenia scanner (Philips Healthcare, Best, The Netherlands). The sequence was evaluated and compared against SShTSE with and without SPAIR in the abdomen of 2 normal volunteers and 2 patients with IRB approval and written informed consent. The typical imaging parameters of the dual-echo Dixon SShTSE sequence included: coronal and/or axial orientation; FOV = 360×360×199 mm; Resolution = 1.5×2×4 mm; SENSE = 3; TE/TR = 45/480 ms, Half Scan factor = 0.6, echo spacing of 6.1ms, δt = 1.1 ms, partial readout of 0.8-1.0. Approximately 40 slices were acquired in 2 breath held acquisitions of 15-17 seconds each. SShTSE with and without SPAIR used similar parameters, except TI = 240 ms for SPAIR, echo spacing of 5.4 ms.Figure 2 shows SShTSE images of a normal volunteer acquired without (a) and with SPAIR (b) in two breath-held acquisitions each. Corresponding images acquired with dual-echo Dixon SShTSE (fig. 2c-e) show uniform fat suppression throughout the volume in two breath-held acquisitions. Similar acquisitions in the axial orientation of a patient at two separate locations are shown in figures 3 and 4 respectively. At both locations, dual-echo Dixon SShTSE provided uniform fat suppression compared to SShTSE with SPAIR.In this work, we developed a true single shot acquisition with uniform fat/water separation by combining dual-echo Dixon acquisition with SShTSE. Optimal combination of partial frequency and phase encoding were implemented to achieve uniform fat/water separation without sacrificing the image quality. This provides a robust SShTSE acquisition with uniform fat/water separation for rapid abdominal imaging.