Hokkaido University Collection of Scholarly and Academic Papers >
Hokkaido University Hospital >
Peer-reviewed Journal Articles, etc >
Development of three-dimensional MR neurography using an optimized combination of compressed sensing and parallel imaging
This item is licensed under:Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International
Title: | Development of three-dimensional MR neurography using an optimized combination of compressed sensing and parallel imaging |
Authors: | Aoike, Takuya Browse this author | Fujima, Noriyuki Browse this author →KAKEN DB | Yoneyama, Masami Browse this author | Fujiwara, Taro Browse this author | Takamori, Sayaka Browse this author | Aoike, Suzuko Browse this author | Ishizaka, Kinya Browse this author | Kudo, Kohsuke Browse this author →KAKEN DB |
Keywords: | Magnetic resonance neurography | Compressed sensing | Sensitivity encoding | Cervical nerve |
Issue Date: | 1-Apr-2022 |
Publisher: | Elsevier |
Journal Title: | Magnetic resonance imaging |
Volume: | 87 |
Start Page: | 32 |
End Page: | 37 |
Publisher DOI: | 10.1016/j.mri.2021.12.002 |
Abstract: | Purpose: To assess the cervical magnetic resonance neurography (MRN) imaging quality obtained with compressed sensing and sensitivity-encoding (compressed SENSE; CS-SENSE) technique in comparison to that obtained with the conventional parallel imaging (i.e., SENSE) technique. Materials and methods: Five healthy volunteers underwent a three-dimensional (3D) turbo spin-echo (TSE)-based cervical MRN examination using a 3.0 Tesla MR-unit. All MRN acquisitions were performed with CS-SENSE and conventional SENSE. We used four acceleration factors (4, 8, 16 and 32) in CS-SENSE. The image quality in MRN was evaluated by assessing the degree of cervical nerve depiction using the contrast ratio (CR) and contrast-noise ratio (CNR) between the cervical nerve and the background signal intensity and a visual scoring system (1: poor, 2: moderate, 3: good). In all of the CR, CNR and visual score, we calculated the ratio of the CS-SENSE-based MRN to that from SENSE-based MRN plus the 95% confidence intervals (CIs) of these ratios. Results: In the multiple comparison of MRN images with the control of conventional SENSE-based MRN, both the quantitative CR values and the visual score for the CS-SENSE factors of 16 and 32 were significantly lower, whereas the CS-SENSE factors of 4 and 8 showed a non-significant difference. In addition, the quantitative CNR values obtained with the CS-SENSE factors of 4 and 8 were significantly higher than that obtained with the conventional SENSE-based MRN while the CS-SENSE factor of 32 was significantly lower, in contrast, the CS SENSE factors of 16 showed a non-significant difference. For CS-SENSE factors of 4 and 8, all ratios of the CS SENSE-based MRN values for CR, CNR and visual scores to those from SENSE-based MRN were above 0.95. Conclusion: CS-SENSE-based MRN can accomplish fast scanning with sufficient image quality when using a high acceleration factor. |
Rights: | © 2022. This manuscript version is made available under the CC-BY-NC-ND 4.0 license https://creativecommons.org/licenses/by-nc-nd/4.0/ | https://creativecommons.org/licenses/by-nc-nd/4.0/ |
Type: | article (author version) |
URI: | http://hdl.handle.net/2115/88788 |
Appears in Collections: | 北海道大学病院 (Hokkaido University Hospital) > 雑誌発表論文等 (Peer-reviewed Journal Articles, etc)
|
Submitter: 藤間 憲幸
|