Abstract
BACKGROUND AND PURPOSE: Synthetic FLAIR images are of lower quality than conventional FLAIR images. Here, we aimed to improve the synthetic FLAIR image quality using deep learning with pixel-by-pixel translation through conditional generative adversarial network training.
MATERIALS AND METHODS: Forty patients with MS were prospectively included and scanned (3T) to acquire synthetic MR imaging and conventional FLAIR images. Synthetic FLAIR images were created with the SyMRI software. Acquired data were divided into 30 training and 10 test datasets. A conditional generative adversarial network was trained to generate improved FLAIR images from raw synthetic MR imaging data using conventional FLAIR images as targets. The peak signal-to-noise ratio, normalized root mean square error, and the Dice index of MS lesion maps were calculated for synthetic and deep learning FLAIR images against conventional FLAIR images, respectively. Lesion conspicuity and the existence of artifacts were visually assessed.
RESULTS: The peak signal-to-noise ratio and normalized root mean square error were significantly higher and lower, respectively, in generated-versus-synthetic FLAIR images in aggregate intracranial tissues and all tissue segments (all P < .001). The Dice index of lesion maps and visual lesion conspicuity were comparable between generated and synthetic FLAIR images (P = 1 and .59, respectively). Generated FLAIR images showed fewer granular artifacts (P = .003) and swelling artifacts (in all cases) than synthetic FLAIR images.
CONCLUSIONS: Using deep learning, we improved the synthetic FLAIR image quality by generating FLAIR images that have contrast closer to that of conventional FLAIR images and fewer granular and swelling artifacts, while preserving the lesion contrast.
ABBREVIATIONS:
- cGAN
- conditional generative adversarial network
- DL
- deep learning
- GAN
- generative adversarial network
- NRMSE
- normalized root mean square error
- PSNR
- peak signal-to-noise ratio
Footnotes
Disclosures: Akifumi Hagiwara—RELATED: Grant: Japan Society for the Promotion of Science KAKENHI, ABiS, Japanese Society for Magnetic Resonance in Medicine, Impulsing Paradigm Change through Disruptive Technologies, Agency for Medical Research and Development, UNRELATED: Payment for Lectures Including Service on Speakers Bureaus: GE Healthcare, Comments: luncheon seminar. Yujiro Otsuka—UNRELATED: Employment: Milliman Inc, Comments: I am receiving a salary as an employee. Yasuhiko Tachibana—UNRELATED: Employment: National Institute of Radiological Sciences, QST; Grants/Grants Pending: Grant-in-Aid for Scientific Research, from the Japan Society for the Promotion of Science, Comments: KAKENHI No. 17K10385. This grant is for a research in the same field as this article but is unrelated to the current study. Koji Kamagata—UNRELATED: Grants/Grants Pending: Japan Society for the Promotion of Science KAKENHI (JP16K19854). Shigeki Aoki—RELATED: Grant: Japan Society for the Promotion of Science KAKENHI, ABiS, Japanese Society for Magnetic Resonance in Medicine, Impulsing Paradigm Change through Disruptive Technologies, Agency for Medical Research and Development, UNRELATED: Grants/Grants Pending: Nihon Medi-Physics, Toshiba, Bayer Yakuhin, Daiichi Sankyo, Eisai, Fujiyakuhin, FUJIFILM RI Pharma, Toshiba, Canon, Japan Society for the Promotion of Science KAKENHI, ABiS, Japanese Society for Magnetic Resonance in Medicine, Impulsing Paradigm Change through Disruptive Technologies, Agency for Medical Research and Development, Payment for Lectures Including Service on Speakers Bureaus: honorarium for lectures/chair from GE Healthcare, Toshiba, Hitachi, Siemens, Bayer Yakuhin, Daiichi Sankyo, Eisai, Fujiyakuhin, FUJIFILM RI Pharma, Nihon Medi-Physics, Meiji Seika Pharma, Canon, Guerbet. Mariko Takemura—RELATED: Grant: Japan Society for the Promotion of Science KAKENHI, Comments: grant No. 16K10327.
This work was supported by Japan Society for the Promotion of Science KAKENHI grant No. 16K19852; grant No. 16K10327; grant No. JP16H06280, Grant-in-Aid for Scientific Research on Innovative Areas, resource and technical support platforms for promoting research “Advanced Bioimaging Support”; the Japanese Society for Magnetic Resonance in Medicine; the Impulsing Paradigm Change through Disruptive Technologies (ImPACT) Program of the Council for Science, Technology and Innovation (cabinet office, Government of Japan); the program for Brain Mapping by Integrated Neurotechnologies for Disease Studies (Brain/MINDS) from the Japan Agency for Medical Research and Development (AMED); and AMED under grant number JP18lk1010025.
- © 2019 by American Journal of Neuroradiology
Indicates open access to non-subscribers at www.ajnr.org