# Deep Learning for Isotropic Super-Resolution from Non-Isotropic 3D Electron Microscopy

## TL;DR

This paper adapts deep learning architectures, specifically FSRCNN and 3D U-Net, to generate 3D isotropic super-resolution from non-isotropic electron microscopy data, demonstrating the superiority of U-Net.

## Contribution

It introduces the use of 3D U-Net for super-resolution in electron microscopy, comparing its performance with FSRCNN, and shows U-Net's consistent better results.

## Key findings

- Both architectures successfully generate 3D isotropic super-resolution.
- U-Net outperforms FSRCNN in all tested scenarios.
- The methods show promise for practical applications in electron microscopy.

## Abstract

The most sophisticated existing methods to generate 3D isotropic super-resolution (SR) from non-isotropic electron microscopy (EM) are based on learned dictionaries. Unfortunately, none of the existing methods generate practically satisfying results. For 2D natural images, recently developed super-resolution methods that use deep learning have been shown to significantly outperform the previous state of the art. We have adapted one of the most successful architectures (FSRCNN) for 3D super-resolution, and compared its performance to a 3D U-Net architecture that has not been used previously to generate super-resolution. We trained both architectures on artificially downscaled isotropic ground truth from focused ion beam milling scanning EM (FIB-SEM) and tested the performance for various hyperparameter settings. Our results indicate that both architectures can successfully generate 3D isotropic super-resolution from non-isotropic EM, with the U-Net performing consistently better. We propose several promising directions for practical application.