Shape-consistent Generative Adversarial Networks for multi-modal Medical segmentation maps
This addresses the challenge of limited data for medical image segmentation, particularly in cardiac scans, but is incremental as it builds on existing GAN and augmentation techniques.
The paper tackled the problem of semantic segmentation in medical imaging with extremely limited datasets by using a 3D cross-modality GAN to synthesize cardiac volumes between CT and MRI, achieving improved segmentation results over other methods with only 16 CT and 16 MRI volumes.
Image translation across domains for unpaired datasets has gained interest and great improvement lately. In medical imaging, there are multiple imaging modalities, with very different characteristics. Our goal is to use cross-modality adaptation between CT and MRI whole cardiac scans for semantic segmentation. We present a segmentation network using synthesised cardiac volumes for extremely limited datasets. Our solution is based on a 3D cross-modality generative adversarial network to share information between modalities and generate synthesized data using unpaired datasets. Our network utilizes semantic segmentation to improve generator shape consistency, thus creating more realistic synthesised volumes to be used when re-training the segmentation network. We show that improved segmentation can be achieved on small datasets when using spatial augmentations to improve a generative adversarial network. These augmentations improve the generator capabilities, thus enhancing the performance of the Segmentor. Using only 16 CT and 16 MRI cardiovascular volumes, improved results are shown over other segmentation methods while using the suggested architecture.