Learning Implicit Brain MRI Manifolds with Deep Learning

TL;DR

This paper demonstrates how deep learning models can learn implicit low-dimensional manifolds of brain MRI data, enabling high-quality image synthesis and denoising to improve neuroimaging analysis.

Contribution

It introduces unsupervised MRI synthesis using GANs and a skip-connected autoencoder for denoising, advancing manifold learning without explicit similarity assumptions.

Findings

Synthesized images are unique and comparable in quality to real images.

Autoencoder denoising outperforms traditional methods in PSNR.

Deep learning effectively models brain MRI manifolds for analysis.

Abstract

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 lowdimensional 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 crosscorrelation 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.