Name
Title | ||
|---|---|---|
Learning How To Interpolate Fourier Data With Unknown Autoregressive Structure: An Ensemble-Based Approach |
Abstract | ||
|---|---|---|
It has been previously shown that the Fourier samples acquired in magnetic resonance imaging (MRI) experiments possess shift-invariant autoregressive structure, which has led to the emergence of various autocalibrated convolution-based image reconstruction approaches. Such approaches, which include GRAPPA, AC-LORAKS, RAKI, and LORAKI, each have their own relative strengths and weaknesses. In this work, we propose a novel ensemble-based approach that uses all of these approaches simultaneously as parallel building blocks within a larger data-adaptive reconstruction network. Results with real data suggest that the ensemble-based approach can synergistically utilize the strengths of each method, providing robust reconstruction performance without the need for interactive parameter tuning. |
Year | DOI | Venue |
|---|---|---|
2019 | 10.1109/IEEECONF44664.2019.9048755 | CONFERENCE RECORD OF THE 2019 FIFTY-THIRD ASILOMAR CONFERENCE ON SIGNALS, SYSTEMS & COMPUTERS |
Keywords | DocType | ISSN |
Computational Imaging, Magnetic Resonance Imaging, Artificial Neural Networks, Autoregression | Conference | 1058-6393 |
Citations | PageRank | References |
0 | 0.34 | 0 |
Authors | ||
2 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Tae Hyung Kim | 1 | 3 | 1.39 |
| Justin P. Haldar | 2 | 350 | 35.40 |