Learning Implicit Brain MRI Manifolds with Deep Learning

• Published in: Proceedings of SPIE, the international society for optical engineering Vol. 10574
• Main Authors: Bermudez, Camilo, Plassard, Andrew J, Davis, Taylor L, Newton, Allen T, Resnick, Susan M, Landman, Bennett A
• Format: Journal Article
• Language: English
• Published: United States 01.03.2018
• Subjects: generative adversarial networks; Manifold learning; brain MRI; image synthesis; deep neural networks
• ISSN: 0277-786X
• DOI: 10.1117/12.2293515

An important task in image processing and neuroimaging is to extract quantitative information from the acquired images in order to make observations about the presence of disease or markers of development in populations. Having a low-dimensional manifold of an image allows for easier statistical comparisons between groups and the synthesis of group representatives. Previous studies have sought to identify the best mapping of brain MRI to a low-dimensional manifold, but have been limited by assumptions of explicit similarity measures. In this work, we use deep learning techniques to investigate implicit manifolds of normal brains and generate new, high-quality images. We explore implicit manifolds by addressing the problems of image synthesis and image denoising as important tools in manifold learning. First, we propose the unsupervised synthesis of T1-weighted brain MRI using a Generative Adversarial Network (GAN) by learning from 528 examples of 2D axial slices of brain MRI. Synthesized images were first shown to be unique by performing a cross-correlation with the training set. Real and synthesized images were then assessed in a blinded manner by two imaging experts providing an image quality score of 1-5. The quality score of the synthetic image showed substantial overlap with that of the real images. Moreover, we use an autoencoder with skip connections for image denoising, showing that the proposed method results in higher PSNR than FSL SUSAN after denoising. This work shows the power of artificial networks to synthesize realistic imaging data, which can be used to improve image processing techniques and provide a quantitative framework to structural changes in the brain.