Key points are not available for this paper at this time.
MRI systems are usually engineered to give 'ideal' performance, and acquisition methods are generally developed using this assumption. This work proposes an alternative sequence learning framework that includes a model of realistic scanner performance, allowing acquisition sequences to be designed to directly account for system imperfections. In this proof-of-concept demonstration we designed pulse sequences to account for eddy current perturbations with different realistic time constants, while also respecting hardware limits. The flexibility of this approach could be used to design new types of pulse sequence to operate on lower performance, lower cost hardware in future.
West et al. (Wed,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: