Diffeomorphic unsupervised deep learning model for mono- and multi-modality registration

Abstract

Different from image segmentation, developing a deep learning network for image registration is less straightforward because training data cannot be prepared or supervised by humans unless they are trivial (e.g. pre-designed affine transforms). One approach for an unsupervised deep leaning model is to self-train the deformation fields by a network based on a loss function with an image similarity metric and a regularisation term, just with traditional variational methods. Such a function consists in a smoothing constraint on the derivatives and a constraint on the determinant of the transformation in order to obtain a spatially smooth and plausible solution. Although any variational model may be used to work with a deep learning algorithm, the challenge lies in achieving robustness. The proposed algorithm is first trained based on a new and robust variational model and tested on synthetic and real mono-modal images. The results show how it deals with large deformation registration problems and leads to a real time solution with no folding. It is then generalised to multi-modal images. Experiments and comparisons with learning and non-learning models demonstrate that this approach can deliver good performances and simultaneously generate an accurate diffeomorphic transformation.

An Unsupervised Deep Learning Method for CT and MR Registration in Spine Surgery Simulator

“…The experimental results showed increase in image registration accuracy in NCC from 0.403 to 0.567. Theljani and Chen [9] proposed an unsupervised deep learning algorithm to achieve the robustness in large deformation registration problem with real time solution. The algorithm was first trained and tested with 2D synthetic and real mono-modal images and then generalized to multi-modal images.…”