Generating Diffusion MRI Scalar Maps from T1 Weighted Images Using Generative Adversarial Networks
(2019) 21st Scandinavian Conference on Image Analysis, SCIA 2019 In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) 11482 LNCS. p.489-498- Abstract
Diffusion magnetic resonance imaging (diffusion MRI) is a non-invasive microstructure assessment technique. Scalar measures, such as FA (fractional anisotropy) and MD (mean diffusivity), quantifying micro-structural tissue properties can be obtained using diffusion models and data processing pipelines. However, it is costly and time consuming to collect high quality diffusion data. Here, we therefore demonstrate how Generative Adversarial Networks (GANs) can be used to generate synthetic diffusion scalar measures from structural T1-weighted images in a single optimized step. Specifically, we train the popular CycleGAN model to learn to map a T1 image to FA or MD, and vice versa. As an application, we show that synthetic FA images can be... (More)
Diffusion magnetic resonance imaging (diffusion MRI) is a non-invasive microstructure assessment technique. Scalar measures, such as FA (fractional anisotropy) and MD (mean diffusivity), quantifying micro-structural tissue properties can be obtained using diffusion models and data processing pipelines. However, it is costly and time consuming to collect high quality diffusion data. Here, we therefore demonstrate how Generative Adversarial Networks (GANs) can be used to generate synthetic diffusion scalar measures from structural T1-weighted images in a single optimized step. Specifically, we train the popular CycleGAN model to learn to map a T1 image to FA or MD, and vice versa. As an application, we show that synthetic FA images can be used as a target for non-linear registration, to correct for geometric distortions common in diffusion MRI.
(Less)
- author
- Gu, Xuan ; Knutsson, Hans LU ; Nilsson, Markus LU and Eklund, Anders
- organization
- publishing date
- 2019
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- keywords
- CycleGAN, Diffusion MRI, Distortion correction, Generative Adversarial Networks
- host publication
- Image Analysis - 21st Scandinavian Conference, SCIA 2019, Proceedings
- series title
- Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
- editor
- Felsberg, Michael ; Forssén, Per-Erik ; Unger, Jonas and Sintorn, Ida-Maria
- volume
- 11482 LNCS
- pages
- 10 pages
- publisher
- Springer
- conference name
- 21st Scandinavian Conference on Image Analysis, SCIA 2019
- conference location
- Norrköping, Sweden
- conference dates
- 2019-06-11 - 2019-06-13
- external identifiers
-
- scopus:85066885746
- ISSN
- 0302-9743
- 1611-3349
- ISBN
- 9783030202040
- DOI
- 10.1007/978-3-030-20205-7_40
- language
- English
- LU publication?
- yes
- id
- 56a9b830-8641-41d1-a7b1-3b87fea91b07
- date added to LUP
- 2019-06-25 13:15:35
- date last changed
- 2024-09-18 02:38:17
@inproceedings{56a9b830-8641-41d1-a7b1-3b87fea91b07, abstract = {{<p>Diffusion magnetic resonance imaging (diffusion MRI) is a non-invasive microstructure assessment technique. Scalar measures, such as FA (fractional anisotropy) and MD (mean diffusivity), quantifying micro-structural tissue properties can be obtained using diffusion models and data processing pipelines. However, it is costly and time consuming to collect high quality diffusion data. Here, we therefore demonstrate how Generative Adversarial Networks (GANs) can be used to generate synthetic diffusion scalar measures from structural T1-weighted images in a single optimized step. Specifically, we train the popular CycleGAN model to learn to map a T1 image to FA or MD, and vice versa. As an application, we show that synthetic FA images can be used as a target for non-linear registration, to correct for geometric distortions common in diffusion MRI.</p>}}, author = {{Gu, Xuan and Knutsson, Hans and Nilsson, Markus and Eklund, Anders}}, booktitle = {{Image Analysis - 21st Scandinavian Conference, SCIA 2019, Proceedings}}, editor = {{Felsberg, Michael and Forssén, Per-Erik and Unger, Jonas and Sintorn, Ida-Maria}}, isbn = {{9783030202040}}, issn = {{0302-9743}}, keywords = {{CycleGAN; Diffusion MRI; Distortion correction; Generative Adversarial Networks}}, language = {{eng}}, pages = {{489--498}}, publisher = {{Springer}}, series = {{Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)}}, title = {{Generating Diffusion MRI Scalar Maps from T1 Weighted Images Using Generative Adversarial Networks}}, url = {{http://dx.doi.org/10.1007/978-3-030-20205-7_40}}, doi = {{10.1007/978-3-030-20205-7_40}}, volume = {{11482 LNCS}}, year = {{2019}}, }