Product-of-Gaussian-mixture diffusion models for joint nonlinear MRI reconstruction
Laurenz Nagler, Martin Zach, Thomas Pock · 2026
Read Product-of-Gaussian-mixture diffusion models for joint nonlinear MRI reconstruction by Laurenz Nagler, Martin Zach, Thomas Pock (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13241435.txt.
Description
Nagler, Zach, and Pock rebuild MRI scans with a diffusion model small enough to stay interpretable. Most diffusion priors are huge black boxes needing pre-estimated coil maps. Their prior multiplies compact Gaussian-mixture experts instead. Image and coil sensitivities are reconstructed jointly, the coils under a smoothness prior.
The method stayed fast and robust when contrast, anatomy, or scan trajectory shifted. A richer parameterization of the prior then improved both denoising and reconstruction quality. No offline coil calibration is needed, freeing acquisition setups. Nagler's team presents it as a leaner route to generative MRI reconstruction.
Catalog record
| CREATOR | Laurenz Nagler, Martin Zach, Thomas Pock |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 573 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |