Weakly Convex Ridge Regularization for 3D Non-Cartesian MRI Reconstruction
This work addresses the need for stable and efficient reconstruction in accelerated non-Cartesian MRI, providing a principled alternative to deep learning methods that is robust to acquisition changes.
The authors propose a rotation invariant weakly convex ridge regularizer (WCRR) for 3D non-Cartesian MRI reconstruction, achieving performance comparable to state-of-the-art deep learning methods while offering improved computational efficiency and robustness to distribution shifts.
While highly accelerated non-Cartesian acquisition protocols significantly reduce scan time, they often entail long reconstruction delays. Deep learning based reconstruction methods can alleviate this, but often lack stability and robustness to distribution shifts. As an alternative, we train a rotation invariant weakly convex ridge regularizer (WCRR). The resulting variational reconstruction approach is benchmarked against state of the art methods on retrospectively simulated data and (out of distribution) on prospective GoLF SPARKLING and CAIPIRINHA acquisitions. Our approach consistently outperforms widely used baselines and achieves performance comparable to Plug and Play reconstruction with a state of the art 3D DRUNet denoiser, while offering substantially improved computational efficiency and robustness to acquisition changes. In summary, WCRR unifies the strengths of principled variational methods and modern deep learning based approaches.