Unsupervised Super-Resolution: Creating High-Resolution Medical Images from Low-Resolution Anisotropic Examples

TL;DR

This paper introduces a novel learning-based super-resolution method that enhances low-resolution anisotropic medical images without needing high-resolution ground truth, using autoencoders trained solely on anisotropic data.

Contribution

It presents a new approach that leverages autoencoder latent spaces to perform super-resolution without high-resolution training data, applicable to various 3D medical images.

Findings

Outperforms conventional interpolation methods in quantitative metrics.

Produces high-quality finer cardiac structures in qualitative assessments.

Applicable to different anatomies and imaging modalities.

Abstract

Although high resolution isotropic 3D medical images are desired in clinical practice, their acquisition is not always feasible. Instead, lower resolution images are upsampled to higher resolution using conventional interpolation methods. Sophisticated learning-based super-resolution approaches are frequently unavailable in clinical setting, because such methods require training with high-resolution isotropic examples. To address this issue, we propose a learning-based super-resolution approach that can be trained using solely anisotropic images, i.e. without high-resolution ground truth data. The method exploits the latent space, generated by autoencoders trained on anisotropic images, to increase spatial resolution in low-resolution images. The method was trained and evaluated using 100 publicly available cardiac cine MR scans from the Automated Cardiac Diagnosis Challenge (ACDC). The quantitative results show that the proposed method performs better than conventional interpolation methods. Furthermore, the qualitative results indicate that especially finer cardiac structures are synthesized with high quality. The method has the potential to be applied to other anatomies and modalities and can be easily applied to any 3D anisotropic medical image dataset.