Cargando…
Revisiting [Formula: see text]-wavelet compressed-sensing MRI in the era of deep learning
Following their success in numerous imaging and computer vision applications, deep-learning (DL) techniques have emerged as one of the most prominent strategies for accelerated MRI reconstruction. These methods have been shown to outperform conventional regularized methods based on compressed sensin...
Autores principales: | , , , , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
National Academy of Sciences
2022
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9388129/ https://www.ncbi.nlm.nih.gov/pubmed/35939712 http://dx.doi.org/10.1073/pnas.2201062119 |
_version_ | 1784770155686920192 |
---|---|
author | Gu, Hongyi Yaman, Burhaneddin Moeller, Steen Ellermann, Jutta Ugurbil, Kamil Akçakaya, Mehmet |
author_facet | Gu, Hongyi Yaman, Burhaneddin Moeller, Steen Ellermann, Jutta Ugurbil, Kamil Akçakaya, Mehmet |
author_sort | Gu, Hongyi |
collection | PubMed |
description | Following their success in numerous imaging and computer vision applications, deep-learning (DL) techniques have emerged as one of the most prominent strategies for accelerated MRI reconstruction. These methods have been shown to outperform conventional regularized methods based on compressed sensing (CS). However, in most comparisons, CS is implemented with two or three hand-tuned parameters, while DL methods enjoy a plethora of advanced data science tools. In this work, we revisit [Formula: see text]-wavelet CS reconstruction using these modern tools. Using ideas such as algorithm unrolling and advanced optimization methods over large databases that DL algorithms utilize, along with conventional insights from wavelet representations and CS theory, we show that [Formula: see text]-wavelet CS can be fine-tuned to a level close to DL reconstruction for accelerated MRI. The optimized [Formula: see text]-wavelet CS method uses only 128 parameters compared to >500,000 for DL, employs a convex reconstruction at inference time, and performs within <1% of a DL approach that has been used in multiple studies in terms of quantitative quality metrics. |
format | Online Article Text |
id | pubmed-9388129 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2022 |
publisher | National Academy of Sciences |
record_format | MEDLINE/PubMed |
spelling | pubmed-93881292022-08-19 Revisiting [Formula: see text]-wavelet compressed-sensing MRI in the era of deep learning Gu, Hongyi Yaman, Burhaneddin Moeller, Steen Ellermann, Jutta Ugurbil, Kamil Akçakaya, Mehmet Proc Natl Acad Sci U S A Physical Sciences Following their success in numerous imaging and computer vision applications, deep-learning (DL) techniques have emerged as one of the most prominent strategies for accelerated MRI reconstruction. These methods have been shown to outperform conventional regularized methods based on compressed sensing (CS). However, in most comparisons, CS is implemented with two or three hand-tuned parameters, while DL methods enjoy a plethora of advanced data science tools. In this work, we revisit [Formula: see text]-wavelet CS reconstruction using these modern tools. Using ideas such as algorithm unrolling and advanced optimization methods over large databases that DL algorithms utilize, along with conventional insights from wavelet representations and CS theory, we show that [Formula: see text]-wavelet CS can be fine-tuned to a level close to DL reconstruction for accelerated MRI. The optimized [Formula: see text]-wavelet CS method uses only 128 parameters compared to >500,000 for DL, employs a convex reconstruction at inference time, and performs within <1% of a DL approach that has been used in multiple studies in terms of quantitative quality metrics. National Academy of Sciences 2022-08-08 2022-08-16 /pmc/articles/PMC9388129/ /pubmed/35939712 http://dx.doi.org/10.1073/pnas.2201062119 Text en Copyright © 2022 the Author(s). Published by PNAS. https://creativecommons.org/licenses/by-nc-nd/4.0/This open access article is distributed under Creative Commons Attribution-NonCommercial-NoDerivatives License 4.0 (CC BY-NC-ND) (https://creativecommons.org/licenses/by-nc-nd/4.0/) . |
spellingShingle | Physical Sciences Gu, Hongyi Yaman, Burhaneddin Moeller, Steen Ellermann, Jutta Ugurbil, Kamil Akçakaya, Mehmet Revisiting [Formula: see text]-wavelet compressed-sensing MRI in the era of deep learning |
title | Revisiting [Formula: see text]-wavelet compressed-sensing MRI in the era of deep learning |
title_full | Revisiting [Formula: see text]-wavelet compressed-sensing MRI in the era of deep learning |
title_fullStr | Revisiting [Formula: see text]-wavelet compressed-sensing MRI in the era of deep learning |
title_full_unstemmed | Revisiting [Formula: see text]-wavelet compressed-sensing MRI in the era of deep learning |
title_short | Revisiting [Formula: see text]-wavelet compressed-sensing MRI in the era of deep learning |
title_sort | revisiting [formula: see text]-wavelet compressed-sensing mri in the era of deep learning |
topic | Physical Sciences |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9388129/ https://www.ncbi.nlm.nih.gov/pubmed/35939712 http://dx.doi.org/10.1073/pnas.2201062119 |
work_keys_str_mv | AT guhongyi revisitingformulaseetextwaveletcompressedsensingmriintheeraofdeeplearning AT yamanburhaneddin revisitingformulaseetextwaveletcompressedsensingmriintheeraofdeeplearning AT moellersteen revisitingformulaseetextwaveletcompressedsensingmriintheeraofdeeplearning AT ellermannjutta revisitingformulaseetextwaveletcompressedsensingmriintheeraofdeeplearning AT ugurbilkamil revisitingformulaseetextwaveletcompressedsensingmriintheeraofdeeplearning AT akcakayamehmet revisitingformulaseetextwaveletcompressedsensingmriintheeraofdeeplearning |