We propose the Automatic-differentiated Physics-Informed Echo State Network (API-ESN). The network is constrained by physical equations through reservoir’s exact time-derivative, which computed automatic differentiation. As compared to original Network, accuracy of time-derivative increased up seven orders magnitude. This key in chaotic dynamical systems, where errors grow exponentially time. s...

A simple phase error correction technique used for field map estimation with a generally available dual-echo gradient-echo (GRE) sequence is presented. Magnetic field inhomogeneity maps estimated using two separate GRE volume acquisitions at different echo times are prone to dynamic motion errors between acquisitions. By using the dual-echo sequence, the data are collected during two back-to-ba...

Introduction Ultra-short Echo Time (UTE) MRI allows for the detection of short T2 components (T2 of the order of 100 μs) in tissue before they have decayed [1, 2]. This is potentially interesting for a range of applications such as MR imaging of bone, tendons and ligaments, sodium MRI, and other applications such as MR-PET attenuation correction. Until now the UTE method has been based on a Ste...

PURPOSE To develop a distortion correction method for echo planar imaging (EPI) that is able to measure dynamic changes in B0 . THEORY AND METHODS The approach we propose is based on single-echo EPI with a jittering of the echo time between two values for alternate time points. Field maps are calculated between phase images from adjacent volumes and are used to remove distortion from correspo...

With improved B 0 homogeneity along with satisfactory gradient performance at high magnetic fields, snapshot gradient-recalled echo-planar imaging (GRE-EPI) would perform at long echo times (TEs) on the order of T2*, which intrinsically allows obtaining strongly T2*-weighted images with embedded substantial anatomical details in ultrashort time. The aim of this study was to investigate the feas...