GAN Based Medical Image Registration
This work addresses the need for faster medical image registration for clinicians and researchers, though it appears incremental as it builds on existing GAN and deep learning techniques.
The authors tackled the problem of slow iterative image registration by proposing an end-to-end deep learning method using GANs, which directly generates registered images with deformation fields in less than a second, achieving accurate results for multimodal retinal and cardiac MR images.
Conventional approaches to image registration consist of time consuming iterative methods. Most current deep learning (DL) based registration methods extract deep features to use in an iterative setting. We propose an end-to-end DL method for registering multimodal images. Our approach uses generative adversarial networks (GANs) that eliminates the need for time consuming iterative methods, and directly generates the registered image with the deformation field. Appropriate constraints in the GAN cost function produce accurately registered images in less than a second. Experiments demonstrate their accuracy for multimodal retinal and cardiac MR image registration.