Rotation effect is another effect during saturation apart from saturation effect, referring to the process of magnetization perpendicular to the saturation direction ($$$\overrightarrow{\omega_{eff}}$$$, for detail see below). The rotation effect would contaminate CEST signal when saturation time is not sufficiently long(empirically $$$t_{sat}<0.8s$$$), which places constrains on repeatable quantification in practices¹. Traditionally this requires restrictively long saturation time, while the concomitant SAR and long scan time would impose a concern. Although sophisticated quantification model can describe rotation effect², it is unavailable to be applied in reality. In this work, a novel Fixed Angle Single Rotation CEST (FASR-CEST) method is proposed which could effectively diminish the rotation effect even if a short saturation period (<0.5s).Based on Zaiss’es analytical solution of Z value^2,3, process of chemical saturation transfer with saturation pulse amplitude $$$B_{1}$$$ and offset $$$\Delta \omega$$$ could be treated as $$$T_{1\rho}$$$ relaxation along the direction of $$$\overrightarrow{\omega_{eff}}$$$ ($$$\overrightarrow{\omega_{eff}}=(\gamma B_{1}, 0, \Delta \omega)$$$), combined with $$$T_{2\rho}$$$ relaxation perpendicular to the direction of $$$\overrightarrow{\omega_{eff}}$$$ (Fig. 1b). If there is still nonzero magnetization on $$$T_{2\rho}$$$ direction ($$$M_{T_{2\rho}}$$$) at the end of saturation pulse, it has net component in Z axis and the rotation effect takes place. Short saturation time will very likely induce rotation effect. To avoid this, rotation pulses before or after rectangular saturation may be proposed to decrease $$$M_{T_{2\rho}}$$$ or its projection to Z axis^3,4. Ideally, the flip angle of the rotation pulse satisfies $$$\beta=\theta$$$, . Considering the inhomogeneity of both $$$B_{0}$$$ and $$$B_{1}$$$, the direction of $$$\overrightarrow{\omega_{eff}}$$$ is varied. Then, the flip angle β should not always be $$$\beta=\theta$$$ as supposed, which would still induce rotation effect. To overcome the defect, only a rectangular rotation pulse before saturation pulse is used (Single Rotation). The rectangular pulse is used as it can decrease difference between β and θ due to unideal $$$B_{1}$$$. Besides, because of inhomogeneous $$$B_{0}$$$, the rotation angles changes with $$$\Delta \omega$$$. Thus, the interpolation method for $$$B_{0}$$$ correction would no longer be suitable. To overcome this issue, the rotation angle is fixed to a single value according to the offset frequency the researcher is most interested in (Fixed Angle) (Fig. 1a). To compare the signal from FASR-CEST sequence (small $$$t_{sat}$$$) with that from conventional CEST, an analytical quantificationmethod (for details see our another abstract) is used, which converts $$$MTR_{asym} $$$ value to $$${\Delta R}_{1\rho}$$$ or $$${\Delta R}_{1\rho}^{calib}$$$. Firstly, phantoms with different solutes (glucose and glutamate) densities and different relaxation times (adjusted with MnCl₂) were prepared and used to test effects of FASR-CEST sequence. Continuous wave rectangular saturation pulse followed by a SE-EPI sequence was used for CEST Imaging. Conventional CEST sequence with $$$B_{1}=1\mu T$$$, $$$TR=2s$$$, $$$t_{sat}=1.5s$$$ was used first acquired. Then, a FASR-CEST sequence with $$$B_{1}=1\mu T$$$, $$$TR=2s$$$, $$$t_{sat}=0.5s$$$ and fixed angle $$$\beta=14.93^{o}$$$ (corresponding to offset frequency around 1.25ppm) was scanned. As comparison, the same CEST sequence with a reduced saturation period ($$$t_{sat}=0.5s$$$) was also scanned. Frequency offsets were from -6ppm to +6ppm. After phantom experiment, in vivo data was acquired on a healthy subject. Three pairs of TR& for CEST imaging were used: $$$TR=2s, t_{sat}=1.5s$$$; $$$TR=1.5s, t_{sat}=1s$$$; $$$TR=7s, t_{sat}=5s$$$, and FASR-CEST sequence was also scanned with $$$TR=2s, t_{sat}=0.5s$$$ and fixed angle $$$\beta=14.93^{o}$$$ (corresponding to around 1.25ppm). For each experiment, corresponding $$$B_{1}$$$ map, $$$T_{1}$$$ map and $$$T_{2}$$$ map were also acquired for quantification.$$${\Delta R}_{1\rho}^{calib}$$$ urves from two different phantoms and two sequences (CEST with $$$t_{sat}=1.5s$$$ and FASR-CEST) are shown in Fig. 2b. It can be seen that the $$${\Delta R}_{1\rho}^{calib}$$$ value from FASR-CEST around 1.25ppm fits well to that from CEST with $$$t_{sat}=1.5s$$$. Besides $$${\Delta R}_{1\rho}^{calib}$$$ map on 1.25ppm from FASR-CEST is also nearly the same with the reference CEST (Fig. 2c, 2d). However, when a short saturation period $$$t_{sat}=0.5s$$$ was used, rotation effect could be obviously observed in an image with almost no effective contrast (Fig. 2e). The effect of FASR-CEST would be similar in vivo data. (Fig. 3)Reproducible quantification in CEST is hurdled by the rotation effect. In practice, it is often non-practical to have sufficiently long saturation period for the rotation effects to be dissipated. In this work a FASR-CEST sequence was proposed to overcome this constrain, phantom and in vivo experiment demonstrated that even with a short saturation period relatively constant CEST quantification can be obtained under varying imaging parameters. FARS-CEST is easy to implement and also help to maintain a low SAR level. Further optimization of this prototype may help its application to other more general saturation strategies.