LeaRning nonlineAr representatIon and projectIon for faSt constrained MRSI rEconstruction (RAIISE)

Learning and utilizing low-dimensional models for high-dimensional spatiospectral imaging problems is an active research area. We present here a novel method for computationally efficient reconstruction from noisy high-dimensional MR spectroscopic imaging (MRSI) data. The proposed method features (a) a novel strategy that jointly learns a nonlinear low-dimensional representation of high-dimensional spectroscopic signals and a neural-network-based projector to recover the low-dimensional embeddings from noisy/limited data; (b) a joint formulation that integrates the forward spatiospectral encoding model, a constraint exploiting the learned representation, and a complementary spatial constraint; and (c) a highly efficient algorithm enabled by a learned projector within an alternating direction method of multipliers (ADMM) framework, circumventing the computationally expensive network inversion subproblem. The proposed method has been evaluated using simulations and in vivo 31P-MRSI and 1H-MRSI data, demonstrating improved performance over state-of-the-art methods. Computational complexity and algorithm convergence analysis have been performed to offer further insights into the effectiveness of the proposed method.