A 3D medical image registration method based on multi-scale feature fusion

Abstract Deformable medical image registration refers to finding a certain transformation so that the corresponding points of two images can be aligned in space. This has important clinical applications. In this article, we propose an unsupervised end-to-end method. method, fixed and moving are concatenated series input into convolution neural network obtain feature different scales. order improve ability capture global local information, fuse maps The spatial uses deformation field deform image, as realize pairs. We validate our method ABIDE data set compare it with some classic state-of-the-art methods. experimental results show improves accuracy