Training Data Independent Image Registration With GANs Using Transfer Learning And Segmentation Information
This addresses a key limitation in automated medical image analysis by improving generalization for new image types, though it appears incremental as it builds on existing GAN and transfer learning techniques.
The paper tackles the problem of deep learning-based image registration methods being heavily dependent on training data and not generalizing well to new image types, by presenting an approach using GANs with transfer learning and segmentation information, which shows better registration performance on chest X-ray and brain MR images compared to conventional methods.
Registration is an important task in automated medical image analysis. Although deep learning (DL) based image registration methods out perform time consuming conventional approaches, they are heavily dependent on training data and do not generalize well for new images types. We present a DL based approach that can register an image pair which is different from the training images. This is achieved by training generative adversarial networks (GANs) in combination with segmentation information and transfer learning. Experiments on chest Xray and brain MR images show that our method gives better registration performance over conventional methods.