Deep Learning Improves Robustness of Voxelwise Kinetic Modeling for Hyperpolarized Carbon‐13 MRI

Purpose To evaluate whether deep learning improves the robustness of voxelwise kinetic parameter estimation from hyperpolarized (HP) 13 C MRI compared with nonlinear least‐squares (NLLS) fitting. Methods A hybrid neural network (NN) was trained on synthetic pyruvate/lactate time courses generated from an open‐system two‐compartment HP 13 C signal model to estimate the pyruvate‐to‐lactate conversion rate (), vascular‐extravascular exchange rate (), and vascular volume fraction (). NN performance was compared with NLLS across flip‐angle schemes, SNR levels, perturbations in acquisition parameters, and in vivo. Matched‐ratio simulations tested whether model‐estimated () retained information beyond the Lac/Pyr area‐under‐the‐curve ratio, . Results In simulations, NLLS and NN performance were comparable for estimation at high SNR, whereas the NN outperformed NLLS at low SNR and for the weakly identifiable parameters and . In vivo, NN maps were more spatially coherent than NLLS maps: corresponded with , while and corresponded with pyruvate AUC. In matched‐ratio simulations, NLLS‐derived discriminated the metabolic classes better than NN‐derived , although both model‐based estimates retained discriminatory information. Conclusion NLLS is effective for estimation under ideal model‐matched conditions, whereas the NN provides more stable voxelwise maps, especially for weakly identifiable parameters and under low‐SNR or in vivo conditions. Prospective biological or repeatability validation is needed to establish quantitative accuracy.