Model uses application of GANs, image-to-image translations, which can be used to obtain multimodal datasets from a single modality. Two unsupervised GAN types (CycleGAN and UNIT) are evaluated for imageto-image translation of T1- and T2-weighted MR images from dataset provided by the Human Connectome project. The evaluation is based on comparison between generated synthetic MR images to ground truth images. It is shown that the implemented GAN models can synthesize visually realistic MR images (incorrectly labeled as real by a human). The Generators_s model outperforms the other models in all quantitative measurements.
Input variables : Paired Images (scans)
Output Variables : Synthesized MRI
Visit Model : github.com
Additional links : arxiv.org
| Model Category | : | Public |
| Date Published | : | June, 2018 |
| Healthcare Domain | : | Medical Technology |
| Code | : | github.com |
| Medical Imaging |
| Image Processing |
| Image Synthesis |
| Radiology |