Deep learning initialized compressed sensing (Deli-CS) in volumetric spatio-temporal subspace reconstruction
Object Spatio-temporal MRI methods offer rapid whole-brain multi-parametric mapping, yet they are often hindered by prolonged reconstruction times or prohibitively burdensome hardware requirements. The aim of this project is to reduce reconstruction time using deep learning. Materials and methods Th...
| Main Authors: | , , , , , , , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | English |
| Published: |
Springer International Publishing
2025
|
| Online Access: | https://hdl.handle.net/1721.1/158263 |
| _version_ | 1826210097572872192 |
|---|---|
| author | Schauman, S. S. Iyer, Siddharth S. Sandino, Christopher M. Yurt, Mahmut Cao, Xiaozhi Liao, Congyu Ruengchaijatuporn, Natthanan Chatnuntawech, Itthi Tong, Elizabeth Setsompop, Kawin |
| author2 | Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science |
| author_facet | Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science Schauman, S. S. Iyer, Siddharth S. Sandino, Christopher M. Yurt, Mahmut Cao, Xiaozhi Liao, Congyu Ruengchaijatuporn, Natthanan Chatnuntawech, Itthi Tong, Elizabeth Setsompop, Kawin |
| author_sort | Schauman, S. S. |
| collection | MIT |
| description | Object Spatio-temporal MRI methods offer rapid whole-brain multi-parametric mapping, yet they are often hindered by prolonged reconstruction times or prohibitively burdensome hardware requirements. The aim of this project is to reduce reconstruction time using deep learning. Materials and methods This study focuses on accelerating the reconstruction of volumetric multi-axis spiral projection MRF, aiming for whole-brain T1 and T2 mapping, while ensuring a streamlined approach compatible with clinical requirements. To optimize reconstruction time, the traditional method is first revamped with a memory-efficient GPU implementation. Deep Learning Initialized Compressed Sensing (Deli-CS) is then introduced, which initiates iterative reconstruction with a DL-generated seed point, reducing the number of iterations needed for convergence. Results The full reconstruction process for volumetric multi-axis spiral projection MRF is completed in just 20 min compared to over 2 h for the previously published implementation. Comparative analysis demonstrates Deli-CS’s efficiency in expediting iterative reconstruction while maintaining high-quality results. Discussion By offering a rapid warm start to the iterative reconstruction algorithm, this method substantially reduces processing time while preserving reconstruction quality. Its successful implementation paves the way for advanced spatio-temporal MRI techniques, addressing the challenge of extensive reconstruction times and ensuring efficient, high-quality imaging in a streamlined manner. |
| first_indexed | 2025-03-10T12:27:46Z |
| format | Article |
| id | mit-1721.1/158263 |
| institution | Massachusetts Institute of Technology |
| language | English |
| last_indexed | 2025-03-10T12:27:46Z |
| publishDate | 2025 |
| publisher | Springer International Publishing |
| record_format | dspace |
| spelling | mit-1721.1/1582632025-02-25T16:16:03Z Deep learning initialized compressed sensing (Deli-CS) in volumetric spatio-temporal subspace reconstruction Schauman, S. S. Iyer, Siddharth S. Sandino, Christopher M. Yurt, Mahmut Cao, Xiaozhi Liao, Congyu Ruengchaijatuporn, Natthanan Chatnuntawech, Itthi Tong, Elizabeth Setsompop, Kawin Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science Object Spatio-temporal MRI methods offer rapid whole-brain multi-parametric mapping, yet they are often hindered by prolonged reconstruction times or prohibitively burdensome hardware requirements. The aim of this project is to reduce reconstruction time using deep learning. Materials and methods This study focuses on accelerating the reconstruction of volumetric multi-axis spiral projection MRF, aiming for whole-brain T1 and T2 mapping, while ensuring a streamlined approach compatible with clinical requirements. To optimize reconstruction time, the traditional method is first revamped with a memory-efficient GPU implementation. Deep Learning Initialized Compressed Sensing (Deli-CS) is then introduced, which initiates iterative reconstruction with a DL-generated seed point, reducing the number of iterations needed for convergence. Results The full reconstruction process for volumetric multi-axis spiral projection MRF is completed in just 20 min compared to over 2 h for the previously published implementation. Comparative analysis demonstrates Deli-CS’s efficiency in expediting iterative reconstruction while maintaining high-quality results. Discussion By offering a rapid warm start to the iterative reconstruction algorithm, this method substantially reduces processing time while preserving reconstruction quality. Its successful implementation paves the way for advanced spatio-temporal MRI techniques, addressing the challenge of extensive reconstruction times and ensuring efficient, high-quality imaging in a streamlined manner. 2025-02-25T16:16:02Z 2025-02-25T16:16:02Z 2025-02-01 2025-02-13T10:16:29Z Article http://purl.org/eprint/type/JournalArticle https://hdl.handle.net/1721.1/158263 Schauman, S.S., Iyer, S.S., Sandino, C.M. et al. Deep learning initialized compressed sensing (Deli-CS) in volumetric spatio-temporal subspace reconstruction. Magn Reson Mater Phy (2025). PUBLISHER_CC en https://doi.org/10.1007/s10334-024-01222-2 Magnetic Resonance Materials in Physics, Biology and Medicine Creative Commons Attribution https://creativecommons.org/licenses/by/4.0/ The Author(s) application/pdf Springer International Publishing Springer International Publishing |
| spellingShingle | Schauman, S. S. Iyer, Siddharth S. Sandino, Christopher M. Yurt, Mahmut Cao, Xiaozhi Liao, Congyu Ruengchaijatuporn, Natthanan Chatnuntawech, Itthi Tong, Elizabeth Setsompop, Kawin Deep learning initialized compressed sensing (Deli-CS) in volumetric spatio-temporal subspace reconstruction |
| title | Deep learning initialized compressed sensing (Deli-CS) in volumetric spatio-temporal subspace reconstruction |
| title_full | Deep learning initialized compressed sensing (Deli-CS) in volumetric spatio-temporal subspace reconstruction |
| title_fullStr | Deep learning initialized compressed sensing (Deli-CS) in volumetric spatio-temporal subspace reconstruction |
| title_full_unstemmed | Deep learning initialized compressed sensing (Deli-CS) in volumetric spatio-temporal subspace reconstruction |
| title_short | Deep learning initialized compressed sensing (Deli-CS) in volumetric spatio-temporal subspace reconstruction |
| title_sort | deep learning initialized compressed sensing deli cs in volumetric spatio temporal subspace reconstruction |
| url | https://hdl.handle.net/1721.1/158263 |
| work_keys_str_mv | AT schaumanss deeplearninginitializedcompressedsensingdelicsinvolumetricspatiotemporalsubspacereconstruction AT iyersiddharths deeplearninginitializedcompressedsensingdelicsinvolumetricspatiotemporalsubspacereconstruction AT sandinochristopherm deeplearninginitializedcompressedsensingdelicsinvolumetricspatiotemporalsubspacereconstruction AT yurtmahmut deeplearninginitializedcompressedsensingdelicsinvolumetricspatiotemporalsubspacereconstruction AT caoxiaozhi deeplearninginitializedcompressedsensingdelicsinvolumetricspatiotemporalsubspacereconstruction AT liaocongyu deeplearninginitializedcompressedsensingdelicsinvolumetricspatiotemporalsubspacereconstruction AT ruengchaijatupornnatthanan deeplearninginitializedcompressedsensingdelicsinvolumetricspatiotemporalsubspacereconstruction AT chatnuntawechitthi deeplearninginitializedcompressedsensingdelicsinvolumetricspatiotemporalsubspacereconstruction AT tongelizabeth deeplearninginitializedcompressedsensingdelicsinvolumetricspatiotemporalsubspacereconstruction AT setsompopkawin deeplearninginitializedcompressedsensingdelicsinvolumetricspatiotemporalsubspacereconstruction |