2 March 2018 Deformable image registration using convolutional neural networks
Author Affiliations +
Deformable image registration can be time-consuming and often needs extensive parameterization to perform well on a specific application. We present a step towards a registration framework based on a three-dimensional convolutional neural network. The network directly learns transformations between pairs of three-dimensional images. The outputs of the network are three maps for the x, y, and z components of a thin plate spline transformation grid. The network is trained on synthetic random transformations, which are applied to a small set of representative images for the desired application. Training therefore does not require manually annotated ground truth deformation information. The methodology is demonstrated on public data sets of inspiration-expiration lung CT image pairs, which come with annotated corresponding landmarks for evaluation of the registration accuracy. Advantages of this methodology are its fast registration times and its minimal parameterization.
© (2018) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Koen A. J. Eppenhof, Koen A. J. Eppenhof, Maxime W. Lafarge, Maxime W. Lafarge, Pim Moeskops, Pim Moeskops, Mitko Veta, Mitko Veta, Josien P. W. Pluim, Josien P. W. Pluim, } "Deformable image registration using convolutional neural networks", Proc. SPIE 10574, Medical Imaging 2018: Image Processing, 105740S (2 March 2018); doi: 10.1117/12.2292443; https://doi.org/10.1117/12.2292443

Back to Top