Low-dose x-ray tomography through a deep convolutional neural network

AI-Driven Advances in Low-Dose Imaging and Enhancement-A Review

The widespread use of medical imaging techniques such as X-rays and computed tomography (CT) has raised significant concerns regarding ionizing radiation exposure, particularly among vulnerable populations requiring frequent imaging. Achieving a balance between high-quality diagnostic imaging and minimizing radiation exposure remains a fundamental challenge in radiology. Artificial intelligence (AI) has emerged as a transformative solution, enabling low-dose imaging protocols that enhance image quality while significantly reducing radiation doses. This review explores the role of AI-assisted low-dose imaging, particularly in CT, X-ray, and magnetic resonance imaging (MRI), highlighting advancements in deep learning models, convolutional neural networks (CNNs), and other AI-based approaches. These technologies have demonstrated substantial improvements in noise reduction, artifact removal, and real-time optimization of imaging parameters, thereby enhancing diagnostic accuracy while mitigating radiation risks. Additionally, AI has contributed to improved radiology workflow efficiency and cost reduction by minimizing the need for repeat scans. The review also discusses emerging directions in AI-driven medical imaging, including hybrid AI systems that integrate post-processing with real-time data acquisition, personalized imaging protocols tailored to patient characteristics, and the expansion of AI applications to fluoroscopy and positron emission tomography (PET). However, challenges such as model generalizability, regulatory constraints, ethical considerations, and computational requirements must be addressed to facilitate broader clinical adoption. AI-driven low-dose imaging has the potential to revolutionize radiology by enhancing patient safety, optimizing imaging quality, and improving healthcare efficiency, paving the way for a more advanced and sustainable future in medical imaging.

AI-NERD: Elucidation of relaxation dynamics beyond equilibrium through AI-informed X-ray photon correlation spectroscopy

Understanding and interpreting dynamics of functional materials in situ is a grand challenge in physics and materials science due to the difficulty of experimentally probing materials at varied length and time scales. X-ray photon correlation spectroscopy (XPCS) is uniquely well-suited for characterizing materials dynamics over wide-ranging time scales. However, spatial and temporal heterogeneity in material behavior can make interpretation of experimental XPCS data difficult. In this work, we have developed an unsupervised deep learning (DL) framework for automated classification of relaxation dynamics from experimental data without requiring any prior physical knowledge of the system. We demonstrate how this method can be used to accelerate exploration of large datasets to identify samples of interest, and we apply this approach to directly correlate microscopic dynamics with macroscopic properties of a model system. Importantly, this DL framework is material and process agnostic, marking a concrete step towards autonomous materials discovery.

Advanced analysis of disintegrating pharmaceutical compacts using deep learning-based segmentation of time-resolved micro-tomography images

The mechanism governing pharmaceutical tablet disintegration is far from fully understood. Despite the importance of controlling a formulation's disintegration process to maximize the active pharmaceutical ingredient's bioavailability and ensure predictable and consistent release profiles, the current understanding of the process is based on indirect or superficial measurements. Formulation science could, therefore, additionally deepen the understanding of the fundamental physical principles governing disintegration based on direct observations of the process. We aim to help bridge the gap by generating a series of time-resolved X-ray micro-computed tomography (μCT) images capturing volumetric images of a broad range of mini-tablet formulations undergoing disintegration. Automated image segmentation was a prerequisite to overcoming the challenges of analyzing multiple time series of heterogeneous tomographic images at high magnification. We devised and trained a convolutional neural network (CNN) based on the U-Net architecture for autonomous, rapid, and consistent image segmentation. We created our own μCT data reconstruction pipeline and parameterized it to deliver image quality optimal for our CNN-based segmentation. Our approach enabled us to visualize the internal microstructures of the tablets during disintegration and to extract parameters of disintegration kinetics from the time-resolved data. We determine by factor analysis the influence of the different formulation components on the disintegration process in terms of both qualitative and quantitative experimental responses. We relate our findings to known formulation component properties and established experimental results. Our direct imaging approach, enabled by deep learning-based image processing, delivers new insights into the disintegration mechanism of pharmaceutical tablets.

Machine learning assists in increasing the time resolution of X-ray computed tomography applied to mineral precipitation in porous media

Many subsurface engineering technologies or natural processes cause porous medium properties, such as porosity or permeability, to evolve in time. Studying and understanding such processes on the pore scale is strongly aided by visualizing the details of geometric and morphological changes in the pores. For realistic 3D porous media, X-Ray Computed Tomography (XRCT) is the method of choice for visualization. However, the necessary high spatial resolution requires either access to limited high-energy synchrotron facilities or data acquisition times which are considerably longer (e.g. hours) than the time scales of the processes causing the pore geometry change (e.g. minutes). Thus, so far, conventional benchtop XRCT technologies are often too slow to allow for studying dynamic processes. Interrupting experiments for performing XRCT scans is also in many instances no viable approach. We propose a novel workflow for investigating dynamic precipitation processes in porous media systems in 3D using a conventional XRCT technology. Our workflow is based on limiting the data acquisition time by reducing the number of projections and enhancing the lower-quality reconstructed images using machine-learning algorithms trained on images reconstructed from high-quality initial- and final-stage scans. We apply the proposed workflow to induced carbonate precipitation within a porous-media sample of sintered glass-beads. So we were able to increase the temporal resolution sufficiently to study the temporal evolution of the precipitate accumulation using an available benchtop XRCT device.

Small training dataset convolutional neural networks for application-specific super-resolution microscopy

Significance

Machine learning (ML) models based on deep convolutional neural networks have been used to significantly increase microscopy resolution, speed [signal-to-noise ratio (SNR)], and data interpretation. The bottleneck in developing effective ML systems is often the need to acquire large datasets to train the neural network. We demonstrate how adding a "dense encoder-decoder" (DenseED) block can be used to effectively train a neural network that produces super-resolution (SR) images from conventional microscopy diffraction-limited (DL) images trained using a small dataset [15 fields of view (FOVs)].

Aim

The ML helps to retrieve SR information from a DL image when trained with a massive training dataset. The aim of this work is to demonstrate a neural network that estimates SR images from DL images using modifications that enable training with a small dataset.

Approach

We employ "DenseED" blocks in existing SR ML network architectures. DenseED blocks use a dense layer that concatenates features from the previous convolutional layer to the next convolutional layer. DenseED blocks in fully convolutional networks (FCNs) estimate the SR images when trained with a small training dataset (15 FOVs) of human cells from the Widefield2SIM dataset and in fluorescent-labeled fixed bovine pulmonary artery endothelial cells samples.

Results

Conventional ML models without DenseED blocks trained on small datasets fail to accurately estimate SR images while models including the DenseED blocks can. The average peak SNR (PSNR) and resolution improvements achieved by networks containing DenseED blocks are ≈ 3.2    dB and 2 × , respectively. We evaluated various configurations of target image generation methods (e.g., experimentally captured a target and computationally generated target) that are used to train FCNs with and without DenseED blocks and showed that including DenseED blocks in simple FCNs outperforms compared to simple FCNs without DenseED blocks.

Conclusions

DenseED blocks in neural networks show accurate extraction of SR images even if the ML model is trained with a small training dataset of 15 FOVs. This approach shows that microscopy applications can use DenseED blocks to train on smaller datasets that are application-specific imaging platforms and there is promise for applying this to other imaging modalities, such as MRI/x-ray, etc.

Artifact identification in X-ray diffraction data using machine learning methods

In situ synchrotron high-energy X-ray powder diffraction (XRD) is highly utilized by researchers to analyze the crystallographic structures of materials in functional devices (e.g. battery materials) or in complex sample environments (e.g. diamond anvil cells or syntheses reactors). An atomic structure of a material can be identified by its diffraction pattern along with a detailed analysis of the Rietveld refinement which yields rich information on the structure and the material, such as crystallite size, microstrain and defects. For in situ experiments, a series of XRD images is usually collected on the same sample under different conditions (e.g. adiabatic conditions) yielding different states of matter, or is simply collected continuously as a function of time to track the change of a sample during a chemical or physical process. In situ experiments are usually performed with area detectors and collect images composed of diffraction patterns. For an ideal powder, the diffraction pattern should be a series of concentric Debye-Scherrer rings with evenly distributed intensities in each ring. For a realistic sample, one may observe different characteristics other than the typical ring pattern, such as textures or preferred orientations and single-crystal diffraction spots. Textures or preferred orientations usually have several parts of a ring that are more intense than the rest, whereas single-crystal diffraction spots are localized intense spots owing to diffraction of large crystals, typically >10 µm. In this work, an investigation of machine learning methods is presented for fast and reliable identification and separation of the single-crystal diffraction spots in XRD images. The exclusion of artifacts during an XRD image integration process allows a precise analysis of the powder diffraction rings of interest. When it is trained with small subsets of highly diverse datasets, the gradient boosting method can consistently produce high-accuracy results. The method dramatically decreases the amount of time spent identifying and separating single-crystal diffraction spots in comparison with the conventional method.

Machine learning in electron microscopy for advanced nanocharacterization: current developments, available tools and future outlook

In the last few years, electron microscopy has experienced a new methodological paradigm aimed to fix the bottlenecks and overcome the challenges of its analytical workflow. Machine learning and artificial intelligence are answering this call providing powerful resources towards automation, exploration, and development. In this review, we evaluate the state-of-the-art of machine learning applied to electron microscopy (and obliquely, to materials and nano-sciences). We start from the traditional imaging techniques to reach the newest higher-dimensionality ones, also covering the recent advances in spectroscopy and tomography. Additionally, the present review provides a practical guide for microscopists, and in general for material scientists, but not necessarily advanced machine learning practitioners, to straightforwardly apply the offered set of tools to their own research. To conclude, we explore the state-of-the-art of other disciplines with a broader experience in applying artificial intelligence methods to their research (e.g., high-energy physics, astronomy, Earth sciences, and even robotics, videogames, or marketing and finances), in order to narrow down the incoming future of electron microscopy, its challenges and outlook.

A holistic overview of deep learning approach in medical imaging

Medical images are a rich source of invaluable necessary information used by clinicians. Recent technologies have introduced many advancements for exploiting the most of this information and use it to generate better analysis. Deep learning (DL) techniques have been empowered in medical images analysis using computer-assisted imaging contexts and presenting a lot of solutions and improvements while analyzing these images by radiologists and other specialists. In this paper, we present a survey of DL techniques used for variety of tasks along with the different medical image's modalities to provide critical review of the recent developments in this direction. We have organized our paper to provide significant contribution of deep leaning traits and learn its concepts, which is in turn helpful for non-expert in medical society. Then, we present several applications of deep learning (e.g., segmentation, classification, detection, etc.) which are commonly used for clinical purposes for different anatomical site, and we also present the main key terms for DL attributes like basic architecture, data augmentation, transfer learning, and feature selection methods. Medical images as inputs to deep learning architectures will be the mainstream in the coming years, and novel DL techniques are predicted to be the core of medical images analysis. We conclude our paper by addressing some research challenges and the suggested solutions for them found in literature, and also future promises and directions for further developments.

Cycloidal CT with CNN-based sinogram completion and in-scan generation of training data

In x-ray computed tomography (CT), the achievable image resolution is typically limited by several pre-fixed characteristics of the x-ray source and detector. Structuring the x-ray beam using a mask with alternating opaque and transmitting septa can overcome this limit. However, the use of a mask imposes an undersampling problem: to obtain complete datasets, significant lateral sample stepping is needed in addition to the sample rotation, resulting in high x-ray doses and long acquisition times. Cycloidal CT, an alternative scanning scheme by which the sample is rotated and translated simultaneously, can provide high aperture-driven resolution without sample stepping, resulting in a lower radiation dose and faster scans. However, cycloidal sinograms are incomplete and must be restored before tomographic images can be computed. In this work, we demonstrate that high-quality images can be reconstructed by applying the recently proposed Mixed Scale Dense (MS-D) convolutional neural network (CNN) to this task. We also propose a novel training approach by which training data are acquired as part of each scan, thus removing the need for large sets of pre-existing reference data, the acquisition of which is often not practicable or possible. We present results for both simulated datasets and real-world data, showing that the combination of cycloidal CT and machine learning-based data recovery can lead to accurate high-resolution images at a limited dose.

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

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.

Joint ptycho-tomography with deep generative priors

Joint ptycho-tomography is a powerful computational imaging framework to recover the refractive properties of a 3D object while relaxing the requirements for probe overlap that is common in conventional phase retrieval. We use an augmented Lagrangian scheme for formulating the constrained optimization problem and employ an alternating direction method of multipliers (ADMM) for the joint solution. ADMM allows the problem to be split into smaller and computationally more efficient subproblems: ptychographic phase retrieval, tomographic reconstruction, and regularization of the solution. We extend our ADMM framework with plug-and-play (PnP) denoisers by replacing the regularization subproblem with a general denoising operator based on machine learning. While the PnP framework enables integrating such learned priors as denoising operators, tuning of the denoiser prior remains challenging. To overcome this challenge, we propose a denoiser parameter to control the effect of the denoiser and to accelerate the solution. In our simulations, we demonstrate that our proposed framework with parameter tuning and learned priors generates high-quality reconstructions under limited and noisy measurement data.

Monitored Tomographic Reconstruction-An Advanced Tool to Study the 3D Morphology of Nanomaterials

Detailed and accurate three-dimensional (3D) information about the morphology of hierarchically structured materials is derived from multi-scale X-ray computed tomography (XCT) and subsequent 3D data reconstruction. High-resolution X-ray microscopy and nano-XCT are suitable techniques to nondestructively study nanomaterials, including porous or skeleton materials. However, laboratory nano-XCT studies are very time-consuming. To reduce the time-to-data by more than an order of magnitude, we propose taking advantage of a monitored tomographic reconstruction. The benefit of this new protocol for 3D imaging is that the data acquisition for each projection is interspersed by image reconstruction. We demonstrate this new approach for nano-XCT data of a novel transition-metal-based materials system: MoNi4 electrocatalysts anchored on MoO2 cuboids aligned on Ni foam (MoNi4/MoO2@Ni). Quantitative data that describe the 3D morphology of this hierarchically structured system with an advanced electrocatalytically active nanomaterial are needed to tailor performance and durability of the electrocatalyst system. We present the framework for monitored tomographic reconstruction, construct three stopping rules for various reconstruction quality metrics and provide their experimental evaluation.

Towards Automatic Detection of Precipitates in Inconel 625 Superalloy Additively Manufactured by the L-PBF Method

In our study, the comparison of the automatically detected precipitates in L-PBF Inconel 625, with experimentally detected phases and with the results of the thermodynamic modeling was used to test their compliance. The combination of the complementary electron microscopy techniques with the microanalysis of chemical composition allowed us to examine the structure and chemical composition of related features. The possibility of automatic detection and identification of precipitated phases based on the STEM-EDS data was presented and discussed. The automatic segmentation of images and identifying of distinguishing regions are based on the processing of STEM-EDS data as multispectral images. Image processing methods and statistical tools are applied to maximize an information gain from data with low signal-to-noise ratio, keeping human interactions on a minimal level. The proposed algorithm allowed for automatic detection of precipitates and identification of interesting regions in the Inconel 625, while significantly reducing the processing time with acceptable quality of results.

Upscaling X-ray nanoimaging to macroscopic specimens

Upscaling X-ray nanoimaging to macroscopic specimens has the potential for providing insights across multiple length scales, but its feasibility has long been an open question. By combining the imaging requirements and existing proof-of-principle examples in large-specimen preparation, data acquisition and reconstruction algorithms, the authors provide imaging time estimates for howX-ray nanoimaging can be scaled to macroscopic specimens. To arrive at this estimate, a phase contrast imaging model that includes plural scattering effects is used to calculate the required exposure and corresponding radiation dose. The coherent X-ray flux anticipated from upcoming diffraction-limited light sources is then considered. This imaging time estimation is in particular applied to the case of the connectomes of whole mouse brains. To image the connectome of the whole mouse brain, electron microscopy connectomics might require years, whereas optimized X-ray microscopy connectomics could reduce this to one week. Furthermore, this analysis points to challenges that need to be overcome (such as increased X-ray detector frame rate) and opportunities that advances in artificial-intelligence-based 'smart' scanning might provide. While the technical advances required are daunting, it is shown that X-ray microscopy is indeed potentially applicable to nanoimaging of millimetre- or even centimetre-size specimens.

Billion-pixel x-ray camera (BiPC-X)

The continuing improvement in quantum efficiency (above 90% for single visible photons), reduction in noise (below 1 electron per pixel), and shrink in pixel pitch (less than 1 μm) enable billion-pixel x-ray cameras (BiPC-X) based on commercial complementary metal-oxide-semiconductor (CMOS) imaging sensors. We describe BiPC-X designs and prototype construction based on flexible tiling of commercial CMOS imaging sensors with millions of pixels. Device models are given for direct detection of low energy x rays (<10 keV) and indirect detection of higher energies using scintillators. Modified Birks's law is proposed for light yield non-proportionality in scintillators as a function of x-ray energy. Single x-ray sensitivity and spatial resolution have been validated experimentally using a laboratory x-ray source and the Argonne Advanced Photon Source. Possible applications include wide field-of-view or large x-ray aperture measurements in high-temperature plasmas, the state-of-the-art synchrotron, x-ray free electron laser, and pulsed power facilities.

Potentials and caveats of AI in hybrid imaging

State-of-the-art patient management frequently mandates the investigation of both anatomy and physiology of the patients. Hybrid imaging modalities such as the PET/MRI, PET/CT and SPECT/CT have the ability to provide both structural and functional information of the investigated tissues in a single examination. With the introduction of such advanced hardware fusion, new problems arise such as the exceedingly large amount of multi-modality data that requires novel approaches of how to extract a maximum of clinical information from large sets of multi-dimensional imaging data. Artificial intelligence (AI) has emerged as one of the leading technologies that has shown promise in facilitating highly integrative analysis of multi-parametric data. Specifically, the usefulness of AI algorithms in the medical imaging field has been heavily investigated in the realms of (1) image acquisition and reconstruction, (2) post-processing and (3) data mining and modelling. Here, we aim to provide an overview of the challenges encountered in hybrid imaging and discuss how AI algorithms can facilitate potential solutions. In addition, we highlight the pitfalls and challenges in using advanced AI algorithms in the context of hybrid imaging and provide suggestions for building robust AI solutions that enable reproducible and transparent research.

Pushing the temporal resolution in absorption and Zernike phase contrast nanotomography: enabling fast in situ experiments

Hard X-ray nanotomography enables 3D investigations of a wide range of samples with high resolution (<100 nm) with both synchrotron-based and laboratory-based setups. However, the advantage of synchrotron-based setups is the high flux, enabling time resolution, which cannot be achieved at laboratory sources. Here, the nanotomography setup at the imaging beamline P05 at PETRA III is presented, which offers high time resolution not only in absorption but for the first time also in Zernike phase contrast. Two test samples are used to evaluate the image quality in both contrast modalities based on the quantitative analysis of contrast-to-noise ratio (CNR) and spatial resolution. High-quality scans can be recorded in 15 min and fast scans down to 3 min are also possible without significant loss of image quality. At scan times well below 3 min, the CNR values decrease significantly and classical image-filtering techniques reach their limitation. A machine-learning approach shows promising results, enabling acquisition of a full tomography in only 6 s. Overall, the transmission X-ray microscopy instrument offers high temporal resolution in absorption and Zernike phase contrast, enabling in situ experiments at the beamline.

Deep Learning-based Inaccuracy Compensation in Reconstruction of High Resolution XCT Data

While X-ray computed tomography (XCT) is pushed further into the micro- and nanoscale, the limitations of various tool components and object motion become more apparent. For high-resolution XCT, it is necessary but practically difficult to align these tool components with sub-micron precision. The aim is to develop a novel reconstruction methodology that considers unavoidable misalignment and object motion during the data acquisition in order to obtain high-quality three-dimensional images and that is applicable for data recovery from incomplete datasets. A reconstruction software empowered by sophisticated correction modules that autonomously estimates and compensates artefacts using gradient descent and deep learning algorithms has been developed and applied. For motion estimation, a novel computer vision methodology coupled with a deep convolutional neural network approach provides estimates for the object motion by tracking features throughout the adjacent projections. The model is trained using the forward projections of simulated phantoms that consist of several simple geometrical features such as sphere, triangle and rectangular. The feature maps extracted by a neural network are used to detect and to classify features done by a support vector machine. For missing data recovery, a novel deep convolutional neural network is used to infer high-quality reconstruction data from incomplete sets of projections. The forward and back projections of simulated geometric shapes from a range of angular ranges are used to train the model. The model is able to learn the angular dependency based on a limited angle coverage and to propose a new set of projections to suppress artefacts. High-quality three-dimensional images demonstrate that it is possible to effectively suppress artefacts caused by thermomechanical instability of tool components and objects resulting in motion, by center of rotation misalignment and by inaccuracy in the detector position without additional computational efforts. Data recovery from incomplete sets of projections result in directly corrected projections instead of suppressing artefacts in the final reconstructed images. The proposed methodology has been proven and is demonstrated for a ball bearing sample. The reconstruction results are compared to prior corrections and benchmarked with a commercially available reconstruction software. Compared to conventional approaches in XCT imaging and data analysis, the proposed methodology for the generation of high-quality three-dimensional X-ray images is fully autonomous. The methodology presented here has been proven for high-resolution micro-XCT and nano-XCT, however, is applicable for all length scales.

Machine learning and big scientific data

This paper reviews some of the challenges posed by the huge growth of experimental data generated by the new generation of large-scale experiments at UK national facilities at the Rutherford Appleton Laboratory (RAL) site at Harwell near Oxford. Such 'Big Scientific Data' comes from the Diamond Light Source and Electron Microscopy Facilities, the ISIS Neutron and Muon Facility and the UK's Central Laser Facility. Increasingly, scientists are now required to use advanced machine learning and other AI technologies both to automate parts of the data pipeline and to help find new scientific discoveries in the analysis of their data. For commercially important applications, such as object recognition, natural language processing and automatic translation, deep learning has made dramatic breakthroughs. Google's DeepMind has now used the deep learning technology to develop their AlphaFold tool to make predictions for protein folding. Remarkably, it has been able to achieve some spectacular results for this specific scientific problem. Can deep learning be similarly transformative for other scientific problems? After a brief review of some initial applications of machine learning at the RAL, we focus on challenges and opportunities for AI in advancing materials science. Finally, we discuss the importance of developing some realistic machine learning benchmarks using Big Scientific Data coming from several different scientific domains. We conclude with some initial examples of our 'scientific machine learning' benchmark suite and of the research challenges these benchmarks will enable. This article is part of a discussion meeting issue 'Numerical algorithms for high-performance computational science'.

Tomographic reconstruction with a generative adversarial network

This paper presents a deep learning algorithm for tomographic reconstruction (GANrec). The algorithm uses a generative adversarial network (GAN) to solve the inverse of the Radon transform directly. It works for independent sinograms without additional training steps. The GAN has been developed to fit the input sinogram with the model sinogram generated from the predicted reconstruction. Good quality reconstructions can be obtained during the minimization of the fitting errors. The reconstruction is a self-training procedure based on the physics model, instead of on training data. The algorithm showed significant improvements in the reconstruction accuracy, especially for missing-wedge tomography acquired at less than 180° rotational range. It was also validated by reconstructing a missing-wedge X-ray ptychographic tomography (PXCT) data set of a macroporous zeolite particle, for which only 51 projections over 70° could be collected. The GANrec recovered the 3D pore structure with reasonable quality for further analysis. This reconstruction concept can work universally for most of the ill-posed inverse problems if the forward model is well defined, such as phase retrieval of in-line phase-contrast imaging.

Deep learning for prediction of cardiac indices from photoplethysmographic waveform: A virtual database approach

Deep learning methods combined with large datasets have recently shown significant progress in solving several medical tasks. However, collecting and annotating large datasets can be a very cumbersome and expensive task. We tackle these problems with a virtual database approach where training data is generated using computer simulations of related phenomena. Specifically, we concentrate on the following problem: can cardiovascular indices such as aortic elasticity, diastolic and systolic blood pressures, and blood flow from heart be predicted continuously using wearable photoplethysmographic sensors? We simulate the blood flow using a haemodynamic model consisting of the entire human circulation. Repeated evaluation of the simulator allows us to create a database of "virtual subjects" with size that is only limited by available computational resources. Using this database, we train neural networks to predict the cardiac indices from photoplethysmographic signal waveform. We consider two approaches: neural networks based on predefined input features and deep convolutional neural networks taking waveform directly as the input. The performance of the methods is demonstrated using numerical examples, thus carrying out a preliminary assessment of the approaches. The results show improvements in accuracy compared with the previous methods. The improvements are especially significant with indices related to aortic elasticity and maximum blood flow. The proposed approach would provide new means to measure cardiovascular health continuously, for example, with a simple wrist device.

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

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.

Improving electron micrograph signal-to-noise with an atrous convolutional encoder-decoder

We present an atrous convolutional encoder-decoder trained to denoise electron micrographs. It consists of a modified Xception backbone, atrous convoltional spatial pyramid pooling module and a multi-stage decoder. Our neural network was trained end-to-end using 512  ×  512 micrographs created from a large dataset of high-dose ( > 2500 counts per pixel) micrographs with added Poisson noise to emulate low-dose ( ≪  300 counts per pixel) data. It was then fine-tuned for high dose data (200-2500 counts per pixel). Its performance is benchmarked against bilateral, Gaussian, median, total variation, wavelet, and Wiener restoration methods with their default parameters. Our network outperforms their best mean squared error and structural similarity index performances by 24.6% and 9.6% for low doses and by 43.7% and 5.5% for high doses. In both cases, our network's mean squared error has the lowest variance. Source code and links to our high-quality dataset and pre-trained models are available at https://github.com/Jeffrey-Ede/Electron-Micrograph-Denoiser.

Accelerating Neutron Tomography experiments through Artificial Neural Network based reconstruction

Neutron Tomography (NT) is a non-destructive technique to investigate the inner structure of a wide range of objects and, in some cases, provides valuable results in comparison to the more common X-ray imaging techniques. However, NT is time consuming and scanning a set of similar objects during a beamtime leads to data redundancy and long acquisition times. Nowadays NT is unfeasible for quality checking study of large quantities of similar objects. One way to decrease the total scan time is to reduce the number of projections. Analytical reconstruction methods are very fast but under this condition generate streaking artifacts in the reconstructed images. Iterative algorithms generally provide better reconstruction for limited data problems, but at the expense of longer reconstruction time. In this study, we propose the recently introduced Neural Network Filtered Back-Projection (NN-FBP) method to optimize the time usage in NT experiments. Simulated and real neutron data were used to assess the performance of the NN-FBP method as a function of the number of projections. For the first time a machine learning based algorithm is applied and tested for NT image reconstruction problem. We demonstrate that the NN-FBP method can reliably reduce acquisition and reconstruction times and it outperforms conventional reconstruction methods used in NT, providing high image quality for limited datasets.

Deep learning in medical imaging and radiation therapy

The goals of this review paper on deep learning (DL) in medical imaging and radiation therapy are to (a) summarize what has been achieved to date; (b) identify common and unique challenges, and strategies that researchers have taken to address these challenges; and (c) identify some of the promising avenues for the future both in terms of applications as well as technical innovations. We introduce the general principles of DL and convolutional neural networks, survey five major areas of application of DL in medical imaging and radiation therapy, identify common themes, discuss methods for dataset expansion, and conclude by summarizing lessons learned, remaining challenges, and future directions.

Back-propagation neural network-based reconstruction algorithm for diffuse optical tomography

Diffuse optical tomography (DOT) is a promising noninvasive imaging modality and is capable of providing functional characteristics of biological tissue by quantifying optical parameters. The DOT image reconstruction is ill-posed and ill-conditioned, due to the highly diffusive nature of light propagation in biological tissues and limited boundary measurements. The widely used regularization technique for DOT image reconstruction is Tikhonov regularization, which tends to yield oversmoothed and low-quality images containing severe artifacts. It is necessary to accurately choose a regularization parameter for Tikhonov regularization. To overcome these limitations, we develop a noniterative reconstruction method, whereby optical properties are recovered based on a back-propagation neural network (BPNN). We train the parameters of BPNN before DOT image reconstruction based on a set of training data. DOT image reconstruction is achieved by implementing a single evaluation of the trained network. To demonstrate the performance of the proposed algorithm, we compare with the conventional Tikhonov regularization-based reconstruction method. The experimental results demonstrate that image quality and quantitative accuracy of reconstructed optical properties are significantly improved with the proposed algorithm.

Synchrotron Big Data Science

The rapid development of synchrotrons has massively increased the speed at which experiments can be performed, while new techniques have increased the amount of raw data collected during each experiment. While this has created enormous new opportunities, it has also created tremendous challenges for national facilities and users. With the huge increase in data volume, the manual analysis of data is no longer possible. As a result, only a fraction of the data collected during the time- and money-expensive synchrotron beam-time is analyzed and used to deliver new science. Additionally, the lack of an appropriate data analysis environment limits the realization of experiments that generate a large amount of data in a very short period of time. The current lack of automated data analysis pipelines prevents the fine-tuning of beam-time experiments, further reducing their potential usage. These effects, collectively known as the "data deluge," affect synchrotrons in several different ways including fast data collection, available local storage, data management systems, and curation of the data. This review highlights the Big Data strategies adopted nowadays at synchrotrons, documenting this novel and promising hybridization between science and technology, which promise a dramatic increase in the number of scientific discoveries.

Improving Tomographic Reconstruction from Limited Data Using Mixed-Scale Dense Convolutional Neural Networks

In many applications of tomography, the acquired data are limited in one or more ways due to unavoidable experimental constraints. In such cases, popular direct reconstruction algorithms tend to produce inaccurate images, and more accurate iterative algorithms often have prohibitively high computational costs. Using machine learning to improve the image quality of direct algorithms is a recently proposed alternative, for which promising results have been shown. However, previous attempts have focused on using encoder–decoder networks, which have several disadvantages when applied to large tomographic images, preventing wide application in practice. Here, we propose the use of the Mixed-Scale Dense convolutional neural network architecture, which was specifically designed to avoid these disadvantages, to improve tomographic reconstruction from limited data. Results are shown for various types of data limitations and object types, for both simulated data and large-scale real-world experimental data. The results are compared with popular tomographic reconstruction algorithms and machine learning algorithms, showing that Mixed-Scale Dense networks are able to significantly improve reconstruction quality even with severely limited data, and produce more accurate results than existing algorithms.