PurposeThe purpose of this study was to seek improved image quality from accelerated echo planar imaging (EPI) data, particularly at ultrahigh fields. Certain artifacts in EPI reconstructions can be attributed to nonlinear phase differences between data acquired using frequency‐encoding gradients of alternating polarity. These errors appear near regions of local susceptibility gradients and typically cannot be corrected with conventional Nyquist ghost correction (NGC) methods.MethodsWe propose a new reconstruction method that integrates ghost correction into the parallel imaging data recovery process. This is achieved through a pair of generalized autocalibrating partially parallel acquisitions (GRAPPA) kernels that operate directly on the measured EPI data. The proposed dual‐polarity GRAPPA (DPG) method estimates missing k‐space data while simultaneously correcting inherent EPI phase errors.ResultsSimulation results showed that standard NGC is incapable of correcting higher‐order phase errors, whereas the DPG kernel approach successfully removed these errors. The presence of higher‐order phase errors near regions of local susceptibility gradients was demonstrated with in vivo data. DPG reconstructions of in vivo 3T and 7T EPI data acquired near these regions showed a marked improvement over conventional methods.ConclusionThis new parallel imaging method for reconstructing accelerated EPI data shows better resilience to inherent EPI phase errors, resulting in higher image quality in regions where higher‐order EPI phase errors commonly occur. Magn Reson Med 76:32–44, 2016. © 2015 Wiley Periodicals, Inc.