We develop a learning framework for building deformable templates, which play a fundamental role in many image analysis and computational anatomy tasks.
We present a probabilistic generative model and derive an unsupervised learning-based inference algorithm that uses insights from classical registration methods and makes use of recent developments in convolutional neural networks (CNNs).
In contrast to this approach, and building on recent learning-based methods, we formulate registration as a function that maps an input image pair to a deformation field that aligns these images.
#2 best model for Diffeomorphic Medical Image Registration on CUMC12
We define registration as a parametric function, and optimize its parameters given a set of images from a collection of interest.
With the "Autograd Image Registration Laboratory" (AirLab), we introduce an open laboratory for image registration tasks, where the analytic gradients of the objective function are computed automatically and the device where the computations are performed, on a CPU or a GPU, is transparent.
We present recursive cascaded networks, a general architecture that enables learning deep cascades, for deformable image registration.
3D medical image registration is of great clinical importance.
In contrast to existing approaches, our framework combines two registration methods: an affine registration and a vector momentum-parameterized stationary velocity field (vSVF) model.
We found that FAIM is able to maintain both the advantages of higher accuracy and fewer "folding" locations over VoxelMorph, over a range of hyper-parameters (with the same values used for both networks).