High-resolution X-ray nanotomography is a quantitative tool for investigating specimens from a wide range of research areas. However, the quality of the reconstructed tomogram is often obscured by noise and therefore not suitable for automatic segmentation. Filtering methods are often required for a detailed quantitative analysis. However, most filters induce blurring in the reconstructed tomograms. Machine learning (ML) techniques offer a powerful alternative to conventional filtering methods. Self-supervised denoising ML technique can be used in a very efficient way for eliminating noise from nanotomography data, outperforming conventional filters, such as a median filter and a nonlocal means filter without blurring relevant structural features.

Below find a list of selected denoising implementions readily available on the Maxwell cluster, and MDLMA publications touching this area.

Also including are more general purpose utilities for image manipulation and visualization

Applications

  1. denoiseg: Joint Denoising and Segmentation
  2. divnoising: Diversity Denoising with Fully Convolutional Variational Autoencoders
  3. histolab: WSI processing in deep learning pipelines
  4. napari: fast, interactive, multi-dimensional image viewer for Python
  5. noise2inverse: Self-supervised deep convolutional denoising for linear inverse problems in imaging
  6. pyg: PyG is a library built upon PyTorch to easily write and train Graph Neural Networks for a wide range of applications related to structured data.
  7. tomopy: Image reconstruction algorithms for tomography
  8. voxelmorph: general purpose library for learning-based tools for alignment/registration and modelling with deformations

Publications

  1. Reconstruction, processing and analysis of tomography data at the Hereon beamlines P05/P07 at PETRA III (Conference Presentation): Julian P. Moosmann et al., SPIE Optical Engineering + Applications, doi: 10.1117/12.2637973
  2. Machine learning denoising of high-resolution X-ray nanotomography data: Silja Flenner et al., J Synchrotron Radiat, doi: 10.1107/S1600577521011139
  3. Artifacts suppression in biomedical images using a guided filter: Inna Bukreeva et al., Thirteenth International Conference on Machine Vision, doi: 10.1117/12.2587571