Enhancing Image Quality of Knee MRI in Low-Field MRI Using Deep Learning [Presidential Award Proceedings]
This study employed deep learning techniques to enhance the image quality of low-field knee MRI, with the goal of aligning it with high-field standards and visualizing transformation effects via texture analysis. Through training on 3T data (51 individuals) and by validating on 0.4T data (25 individuals) with two distinct models (super-resolution and contrast generation), the results might reveal an enhancement in the image quality of low-field knee MRI.
Authors
- Ryosuke Sakai (ORCID: https://orcid.org/0000-0002-2396-0747)
- Tomoya Nakatsuka
- Akihiko Wada (ORCID: https://orcid.org/0000-0003-3432-0267)
- Tsutomu Inaoka (ORCID: https://orcid.org/0000-0002-5251-077X)
- Masayuki Sugeta (ORCID: https://orcid.org/0000-0002-6722-2260)
- Hitoshi Terada
- Hisanori Tomobe
- Masaru Sonoda
Institutions
- Toho University (JP)
- Juntendo University (JP)
- Toho University Medical Center Sakura Hospital (JP)
- Seirei Sakura Citizen Hospital (JP)
Publication Details
- Journal
- Japanese Journal of Magnetic Resonance in Medicine
- Published
- 2024-08-15
- DOI
- https://doi.org/10.2463/jjmrm.2023-1805
- Citations
- 1
- Primary Topic
- Photoacoustic and Ultrasonic Imaging
- Type
- article
- Field-Weighted Citation Impact
- 0.18