Improving the Resolution and SNR of Diffusion Magnetic Resonance Images From a Low-Field Scanner
(2023) In Lecture Notes in Networks and Systems 746. p.147-160- Abstract
- Spatial resolution, signal-to-noise ratio (SNR) and acquisition time are interconnected in magnetic resonance imaging (MRI). Trade-offs are made to keep the SNR at the acceptable level, maximizing the resolution, minimizing the acquisition time and maintaining radiologically useful images. In low-field MRI scanners and especially in diffusion imaging, these trade-offs are even more crucial due to a generally lower image quality. Image post-processing is necessary in such cases to improve image quality. In this work, we alleviate the challenges of low SNR in dMRI at low magnetic fields by performing super-resolution reconstruction (SRR). Our approach combines multiple low-resolution images acquired at different image slice rotations and... (More)
- Spatial resolution, signal-to-noise ratio (SNR) and acquisition time are interconnected in magnetic resonance imaging (MRI). Trade-offs are made to keep the SNR at the acceptable level, maximizing the resolution, minimizing the acquisition time and maintaining radiologically useful images. In low-field MRI scanners and especially in diffusion imaging, these trade-offs are even more crucial due to a generally lower image quality. Image post-processing is necessary in such cases to improve image quality. In this work, we alleviate the challenges of low SNR in dMRI at low magnetic fields by performing super-resolution reconstruction (SRR). Our approach combines multiple low-resolution images acquired at different image slice rotations and employs a convolutional neural network to perform the SRR. Training is performed on noisy images. The network learns to extract and compose complementary image details into a super-resolution output image. Because of the properties of noise and the training process, the super-resolution images are less noisy than the directly acquired high-resolution ones, contain more high-resolution details than the input low-resolution images and the total acquisition time is decreased. (Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/d2759004-765d-489c-bd53-897b26f6bbb7
- author
- Jurek, Jakub ; Ludwisiak, Kamil ; Materka, Andzej and Szczepankiewicz, Filip LU
- organization
- publishing date
- 2023-09-11
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- host publication
- The Latest Developments and Challenges in Biomedical Engineering : Proceedings of the 23rd Polish Conference on Biocybernetics and Biomedical Engineering, Lodz, Poland, September 27–29, 2023 - Proceedings of the 23rd Polish Conference on Biocybernetics and Biomedical Engineering, Lodz, Poland, September 27–29, 2023
- series title
- Lecture Notes in Networks and Systems
- volume
- 746
- pages
- 147 - 160
- publisher
- Springer Nature
- external identifiers
-
- scopus:85172358441
- ISSN
- 2367-3389
- 2367-3370
- ISBN
- 978-3-031-38430-1
- 978-3-031-38429-5
- DOI
- 10.1007/978-3-031-38430-1_12
- language
- English
- LU publication?
- yes
- id
- d2759004-765d-489c-bd53-897b26f6bbb7
- date added to LUP
- 2023-09-18 09:55:01
- date last changed
- 2024-04-17 09:53:17
@inproceedings{d2759004-765d-489c-bd53-897b26f6bbb7, abstract = {{Spatial resolution, signal-to-noise ratio (SNR) and acquisition time are interconnected in magnetic resonance imaging (MRI). Trade-offs are made to keep the SNR at the acceptable level, maximizing the resolution, minimizing the acquisition time and maintaining radiologically useful images. In low-field MRI scanners and especially in diffusion imaging, these trade-offs are even more crucial due to a generally lower image quality. Image post-processing is necessary in such cases to improve image quality. In this work, we alleviate the challenges of low SNR in dMRI at low magnetic fields by performing super-resolution reconstruction (SRR). Our approach combines multiple low-resolution images acquired at different image slice rotations and employs a convolutional neural network to perform the SRR. Training is performed on noisy images. The network learns to extract and compose complementary image details into a super-resolution output image. Because of the properties of noise and the training process, the super-resolution images are less noisy than the directly acquired high-resolution ones, contain more high-resolution details than the input low-resolution images and the total acquisition time is decreased.}}, author = {{Jurek, Jakub and Ludwisiak, Kamil and Materka, Andzej and Szczepankiewicz, Filip}}, booktitle = {{The Latest Developments and Challenges in Biomedical Engineering : Proceedings of the 23rd Polish Conference on Biocybernetics and Biomedical Engineering, Lodz, Poland, September 27–29, 2023}}, isbn = {{978-3-031-38430-1}}, issn = {{2367-3389}}, language = {{eng}}, month = {{09}}, pages = {{147--160}}, publisher = {{Springer Nature}}, series = {{Lecture Notes in Networks and Systems}}, title = {{Improving the Resolution and SNR of Diffusion Magnetic Resonance Images From a Low-Field Scanner}}, url = {{http://dx.doi.org/10.1007/978-3-031-38430-1_12}}, doi = {{10.1007/978-3-031-38430-1_12}}, volume = {{746}}, year = {{2023}}, }