Machine learning denoising of high-resolution X-ray nanotomography data

Abstract

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. Here, machine learning (ML) techniques offer a powerful alternative to conventional filtering methods. In this article, we verify that a self-supervised denoising ML technique can be used in a very efficient way for eliminating noise from nanotomography data. The technique presented is applied to high-resolution nanotomography data and compared to conventional filters, such as a median filter and a nonlocal means filter, optimized for tomographic data sets. The ML approach proves to be a very powerful tool that outperforms conventional filters by eliminating noise without blurring relevant structural features, thus enabling efficient quantitative analysis in different scientific fields.

Hard X-ray full-field nanoimaging using a direct photon-counting detector

“…However, already in the 36 s scan, single microtrichia are clearly resolved. The noise in this scan could be easily removed using, for example, an iterative non-local means filter (Bruns et al, 2017) or a machine learning based approach (Pelt & Sethian, 2018;Hendriksen et al, 2020;Flenner et al, 2022). The slices obtained from scans with longer acquisition time show a decreasing noise level and a high CNR [Fig.…”

“…Here, a long sample-to-detector distance is needed to achieve a sufficient magnification in the X-ray regime due to the small divergence of the focused X-ray beam. This makes the system more efficient, but the noise level, mainly a combination of detector readout noise and photon noise, is still not negligible (Flenner et al, 2022).…”

Performance Evaluation of Deep Neural Network Model for Coherent X-ray Imaging

“…In the former case, a diffractive image leads a convergence of solution to an inaccurate direction and in the latter case, the prediction is less reliable due to the lack of similarity between the images used in pre-training and actual input image. Therefore, the approach to enhance the accuracy of iterative models is to improve the quality of diffractive images such as denoising images [62] or to improve noise tolerance in phase retrieval process [63,64], whereas the effort to increase the reliability of end-to-end model is to make the images for pre-training similar to the input images fed into the model [56].…”

Transfer Learning Application of Self-supervised Learning in ARPES

“…In short, these models learn the feature from the input data in an selfsupervised way, which means the training objective is generated without using annotated human labels, but using pretext tasks. In the context of natural science experiment, there are already attempts to apply some form of self-supervised learning model, for example in the training of self-supervised model built for single cell image analysis [38], training of selfsupervised model to denoise tomography data (Noise2inverse) [39,40], analysis of electrocardiography (ECG) database with a general purpose self-supervised model [41],…”