[KongchangAI]
ConceptUnrolled Reconstruction Network

Unrolled Reconstruction Network

A class of deep learning methods that unroll traditional iterative optimization algorithms into a finite number of neural network layers, each containing data consistency and learnable regularization modules, combining physical model constraints with end-to-end learning capability.

Timeline (last 90 days)

Nov 17

Unrolled reconstruction networks unroll iterative optimization algorithms into a finite number of neural network layers that alternately perform data consistency and regularization, where regularization is replaced by learnable modules and data consistency retains its analytical form.

Unverified50%
Nov 17

Compared with purely data-driven black-box networks, unrolled networks have fewer parameters, are more interpretable, and are more robust to changes in undersampling patterns, and have become one of the mainstream paradigms for accelerated MRI reconstruction.

Unverified50%

All Facts (2)

Source Articles