TomoGAN: low-dose synchrotron x-ray tomography with generative adversarial networks: discussion

Abstract

Synchrotron-based x-ray tomography is a noninvasive imaging technique that allows for reconstructing the internal structure of materials at high spatial resolutions from tens of micrometers to a few nanometers. In order to resolve sample features at smaller length scales, however, a higher radiation dose is required. Therefore, the limitation on the achievable resolution is set primarily by noise at these length scales. We present TomoGAN, a denoising technique based on generative adversarial networks, for improving the quality of reconstructed images for low-dose imaging conditions. We evaluate our approach in two photon-budget-limited experimental conditions: (1) sufficient number of low-dose projections (based on Nyquist sampling), and (2) insufficient or limited number of high-dose projections. In both cases the angular sampling is assumed to be isotropic, and the photon budget throughout the experiment is fixed based on the maximum allowable radiation dose on the sample. Evaluation with both simulated and experimental datasets shows that our approach can significantly reduce noise in reconstructed images, improving the structural similarity score of simulation and experimental data from 0.18 to 0.9 and from 0.18 to 0.41, respectively. Furthermore, the quality of the reconstructed images with filtered back projection followed by our denoising approach exceeds that of reconstructions with the simultaneous iterative reconstruction technique, showing the computational superiority of our approach.

Deep Learning Accelerated Light Source Experiments

“…We append TomoGAN, an image quality enhancement model based on generative adversarial networks [5] originally developed for low-dose X-ray imaging in [6], to the streaming tomographic processing pipeline to enable online enhancement of image quality. We have shown in previous work [6] that, once trained on one sample, TomoGAN can be applied effectively to other similar samples, even if X-ray projections of those samples are collected at a different facility and show different noise characteristics. Fig.…”

“…Yang et al [48], use a convolutional neural network (CNN) to denoize reconstructed images and show 10-fold improvement on signal-to-noise ratio. In our work, we apply our denoising method, TomoGAN [6], to streaming reconstructions and evaluate its impact on image quality and end-to-end performance.…”

“…• We propose and implement a pipelined workflow for tomography reconstruction with streaming data (as shown in Figure 2) for realtime streaming analysis; • We repurpose and retrain TomoGAN (a GAN [5]-based deep learning model originally designed for low dose Xray tomography [6]) and integrate it into our workflow for image quality enhancement (as shown in Figure 3); • We evaluate our system with two real-world data sets collected at APS and provide insightful analysis on performance improvements.…”

Scientific Image Restoration Anywhere

“…We used two datasets for our experiments. Each dataset comprises 1024 pairs of 1024×1024 images, each pair being a noisy image and a corresponding ground-truth image, as described in Liu et al [15]. Ground truth images are obtained from normal-dose X-ray imaging and noisy images from lowdose X-ray imaging of the same sample.…”

“…We explore these questions here by studying how a specific scientific deep learning model, TomoGAN 1 [15,16], can be adapted for edge deployment. TomoGAN uses generative adversarial network (GAN) methods [17] to enhance the quality of low-dose X-ray images via a denoising process.…”