The approach proposes a deep convolutional adversarial network framework by adversarially training FCN as the generator and CNN as the discriminator to generate target image with high modality using source image with low modality. To avoid generating blurry target images, the FCN is designed to incorporate an image-gradient-difference-based loss function. Further application of Auto-Context Model implements a context-aware deep convolutional adversarial network. The experiments have shown that the method is accurate and robust for synthesizing target images from the corresponding source images.
Input variables : Images with low modality
Output Variables : Synthesized images with high modality
Visit Model : github.com
Additional links : ncbi.nlm.nih.gov
| Model Category | : | Public |
| Date Published | : | December, 2018 |
| Healthcare Domain | : | Medical Technology |
| Code | : | github.com |
| Medical Imaging |
| Image Synthesis |