Enhanced Partial Fourier MRI With Zero-Shot Deep Untrained Priors
SO HYUN KANG, JIHOO KIM, JAEJIN CHO, CLARISSA ZIMMERMAN COOLEY, AMIR HEYDARI, BERKIN BILGIC, TAE HYUNG KIM · 2024
Read Enhanced Partial Fourier MRI With Zero-Shot Deep Untrained Priors by SO HYUN KANG, JIHOO KIM, JAEJIN CHO, CLARISSA ZIMMERMAN COOLEY, AMIR HEYDARI, BERKIN BILGIC, TAE HYUNG KIM (2024) 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-12323813.txt.
Description
So Hyun Kang, Jihoo Kim, and their co-authors present a 2024 method to sharpen MRI scans that are taken with only part of the usual data. Partial scans are faster, but filling in the missing part well is hard. Their method joins a classic phase based rule with a newer deep learning trick that needs no training data at all.
The trick rests on the idea that a fresh, untrained network can itself capture the shape of good images while it rebuilds one. The team tested the combined method on three data sets, including a public brain and knee set and a low field scanner set. Image quality beat older methods, with error cut by factors from about 1.15 to 2.5. Faster scans with clean images could widen MRI access.
Catalog record
| CREATOR | SO HYUN KANG, JIHOO KIM, JAEJIN CHO, CLARISSA ZIMMERMAN COOLEY, AMIR HEYDARI, BERKIN BILGIC, TAE HYUNG KIM |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 30 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |