Purpose An end‐to‐end differentiable 2D Bloch simulation is used to reduce T2 induced blurring in single‐shot turbo spin echo sequences, also called rapid imaging with refocused echoes (RARE) sequences, by using a joint optimization of refocusing flip angles and a convolutional neural network. Methods Simulation and optimization were performed in the MR‐zero framework. Variable flip angle train and DenseNet parameters were optimized jointly using the instantaneous transverse magnetization, available in our simulation, at a certain echo time, which serves as ideal blurring‐free target. Final optimized sequences were exported for in vivo measurements at a real system (3 T Siemens, PRISMA) using the Pulseq standard. Results The optimized RARE was able to successfully lower T2‐induced blurring for single‐shot RARE sequences in proton density‐weighted and T2‐weighted images. In addition to an increased sharpness, the neural network allowed correction of the contrast changes to match the theoretical transversal magnetization. The optimization found flip angle design strategies similar to existing literature, however, visual inspection of the images and evaluation of the respective point spread function demonstrated an improved performance. Conclusions This work demonstrates that when variable flip angles and a convolutional neural network are optimized jointly in an end‐to‐end approach, sequences with more efficient minimization of T2‐induced blurring can be found. This allows faster single‐ or multi‐shot RARE MRI with longer echo trains.

We propose an alternate method for simulating arbitrary MRI sequences, forming complete Bloch simulations. It is closely related to Extended Phase Graphs but imposes no restrictions on the alignment of timing or gradients, extends it to support T2' relaxation and 3D gradient encoding, is fully differentiable, provides additional insight into the composition of the measured signal while outperforming isochromat-based Bloch simulations in execution speed by using an analytical description of the signal instead of relying on a Monte-Carlo simulation.

We propose a method to automatically generate estimators for the exact encoding of a MRI signal, based on the Phase Distribution Graph of a sequence and its simulation. The estimator can then be used on a measurement to split the signal into its differently encoded parts, which enables reconstruction tailored to the sequence used. This approach can result in fewer imaging artifacts since it does not rely on any assumptions about the sequence made up front but on data obtained by a PDG simulation. This also makes it suitable for sequence optimization because it can adapt to changing sequence properties.