MR Slice Profile Estimation by Learning to Match Internal Patch Distributions

To super-resolve the through-plane direction of a multi-slice 2D magnetic resonance (MR) image, its slice selection profile can be used as the degeneration model from high resolution (HR) to low resolution (LR) to create paired data when training a supervised algorithm. Existing super-resolution algorithms make assumptions about the slice selection profile since it is not readily known for a given image. In this work, we estimate a slice selection profile given a specific image by learning to match its internal patch distributions. Specifically, we assume that after applying the correct slice selection profile, the image patch distribution along HR in-plane directions should match the distribution along the LR through-plane direction. Therefore, we incorporate the estimation of a slice selection profile as part of learning a generator in a generative adversarial network (GAN). In this way, the slice selection profile can be learned without any external data. Our algorithm was tested using simulations from isotropic MR images, incorporated in a through-plane super-resolution algorithm to demonstrate its benefits, and also used as a tool to measure image resolution. Our code is at https://github.com/shuohan/espreso2.

[1]  R. Lerski An evaluation using computer simulation of two methods of slice profile determination in MRI , 1989 .

[2]  Dijia Wu,et al.  Reduce slice spacing of MR images by super-resolution learned without ground-truth , 2020 .

[3]  Andrea Vedaldi,et al.  Deep Image Prior , 2017, International Journal of Computer Vision.

[4]  Karin Schwab,et al.  Medical Imaging Signals And Systems , 2016 .

[5]  Yuichi Yoshida,et al.  Spectral Normalization for Generative Adversarial Networks , 2018, ICLR.

[6]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[7]  Haiying Liu,et al.  Actual imaging slice profile of 2D MRI , 2002, SPIE Medical Imaging.

[8]  Qinghua Hu,et al.  Neural Blind Deconvolution Using Deep Priors , 2019, 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[9]  Eugene W. Myers,et al.  Isotropic reconstruction of 3D fluorescence microscopy images using convolutional neural networks , 2017, MICCAI.

[10]  Aaron Carass,et al.  Applications of a deep learning method for anti-aliasing and super-resolution in MRI. , 2019, Magnetic resonance imaging.

[11]  Michal Irani,et al.  Blind Super-Resolution Kernel Estimation using an Internal-GAN , 2019, NeurIPS.

[12]  Yoshua Bengio,et al.  Generative Adversarial Nets , 2014, NIPS.

[13]  Subhransu Maji,et al.  A Bayesian Perspective on the Deep Image Prior , 2019, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[14]  Peter A. Calabresi,et al.  A Deep Learning Based Anti-aliasing Self Super-Resolution Algorithm for MRI , 2018, MICCAI.

[15]  Stefan Winkler,et al.  The Unusual Effectiveness of Averaging in GAN Training , 2018, ICLR.

[16]  Qi Tian,et al.  Unsupervised Image Super-Resolution with an Indirect Supervised Path , 2019, 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).

[17]  Daniel S. Marcus,et al.  OASIS-3: Longitudinal Neuroimaging, Clinical, and Cognitive Dataset for Normal Aging and Alzheimer Disease , 2019 .

[18]  Dong Liu,et al.  Isotropic Reconstruction of 3D EM Images with Unsupervised Degradation Learning , 2020, MICCAI.