Deploying Machine Learning Based Segmentation for Scientific Imaging Analysis at Synchrotron Facilities.

Scientific user facilities present a unique set of challenges for image processing due to the large volume of data generated from experiments and simulations. Furthermore, developing and implementing algorithms for real-time processing and analysis while correcting for any artifacts or distortions in images remains a complex task, given the computational requirements of the processing algorithms. In a collaborative effort across multiple Department of Energy national laboratories, the "MLExchange" project is focused on addressing these challenges. MLExchange is a Machine Learning framework deploying interactive web interfaces to enhance and accelerate data analysis. The platform allows users to easily upload, visualize, label, and train networks. The resulting models can be deployed on real data while both results and models could be shared with the scientists. The MLExchange web-based application for image segmentation allows for training, testing, and evaluating multiple machine learning models on hand-labeled tomography data. This environment provides users with an intuitive interface for segmenting images using a variety of machine learning algorithms and deep-learning neural networks. Additionally, these tools have the potential to overcome limitations in traditional image segmentation techniques, particularly for complex and low-contrast images.

The AXEAP2 program for Kß X-ray emission spectra analysis using artificial intelligence.

The processing and analysis of synchrotron data can be a complex task, requiring specialized expertise and knowledge. Our previous work addressed the challenge of X-ray emission spectrum (XES) data processing by developing a standalone application using unsupervised machine learning. However, the task of analyzing the processed spectra remains another challenge. Although the non-resonant Kß XES of 3d transition metals are known to provide electronic structure information such as oxidation and spin state, finding appropriate parameters to match experimental data is a time-consuming and labor-intensive process. Here, a new XES data analysis method based on the genetic algorithm is demonstrated, applying it to Mn, Co and Ni oxides. This approach is also implemented as a standalone application, Argonne X-ray Emission Analysis 2 (AXEAP2), which finds a set of parameters that result in a high-quality fit of the experimental spectrum with minimal intervention. AXEAP2 is able to find a set of parameters that reproduce the experimental spectrum, and provide insights into the 3d electron spin state, 3d-3p electron exchange force and Kß emission core-hole lifetime.

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.

Linking scientific instruments and computation: Patterns, technologies, and experiences.

Powerful detectors at modern experimental facilities routinely collect data at multiple GB/s. Online analysis methods are needed to enable the collection of only interesting subsets of such massive data streams, such as by explicitly discarding some data elements or by directing instruments to relevant areas of experimental space. Thus, methods are required for configuring and running distributed computing pipelines-what we call flows-that link instruments, computers (e.g., for analysis, simulation, artificial intelligence [AI] model training), edge computing (e.g., for analysis), data stores, metadata catalogs, and high-speed networks. We review common patterns associated with such flows and describe methods for instantiating these patterns. We present experiences with the application of these methods to the processing of data from five different scientific instruments, each of which engages powerful computers for data inversion,model training, or other purposes. We also discuss implications of such methods for operators and users of scientific facilities.

AXEAP: a software package for X-ray emission data analysis using unsupervised machine learning.

The Argonne X-ray Emission Analysis Package (AXEAP) has been developed to calibrate and process X-ray emission spectroscopy (XES) data collected with a two-dimensional (2D) position-sensitive detector. AXEAP is designed to convert a 2D XES image into an XES spectrum in real time using both calculations and unsupervised machine learning. AXEAP is capable of making this transformation at a rate similar to data collection, allowing real-time comparisons during data collection, reducing the amount of data stored from gigabyte-sized image files to kilobyte-sized text files. With a user-friendly interface, AXEAP includes data processing for non-resonant and resonant XES images from multiple edges and elements. AXEAP is written in MATLAB and can run on common operating systems, including Linux, Windows, and MacOS.

pyXPCSviewer: an open-source interactive tool for X-ray photon correlation spectroscopy visualization and analysis.

pyXPCSviewer, a Python-based graphical user interface that is deployed at beamline 8-ID-I of the Advanced Photon Source for interactive visualization of XPCS results, is introduced. pyXPCSviewer parses rich X-ray photon correlation spectroscopy (XPCS) results into independent PyQt widgets that are both interactive and easy to maintain. pyXPCSviewer is open-source and is open to customization by the XPCS community for ingestion of diversified data structures and inclusion of novel XPCS techniques, both of which are growing demands particularly with the dawn of near-diffraction-limited synchrotron sources and their dedicated XPCS beamlines.

MLExchange: A web-based platform enabling exchangeable machine learning workflows for scientific studies.

Machine learning (ML) algorithms are showing a growing trend in helping the scientific communities across different disciplines and institutions to address large and diverse data problems. However, many available ML tools are programmatically demanding and computationally costly. The MLExchange project aims to build a collaborative platform equipped with enabling tools that allow scientists and facility users who do not have a profound ML background to use ML and computational resources in scientific discovery. At the high level, we are targeting a full user experience where managing and exchanging ML algorithms, workflows, and data are readily available through web applications. Since each component is an independent container, the whole platform or its individual service(s) can be easily deployed at servers of different scales, ranging from a personal device (laptop, smart phone, etc.) to high performance clusters (HPC) accessed (simultaneously) by many users. Thus, MLExchange renders flexible using scenarios-users could either access the services and resources from a remote server or run the whole platform or its individual service(s) within their local network.

20†µs-resolved high-throughput X-ray photon correlation spectroscopy on a 500k pixel detector enabled by data-management workflow.

The performance of the new 52 kHz frame rate Rigaku XSPA-500k detector was characterized on beamline 8-ID-I at the Advanced Photon Source at Argonne for X-ray photon correlation spectroscopy (XPCS) applications. Due to the large data flow produced by this detector (0.2 PB of data per 24 h of continuous operation), a workflow system was deployed that uses the Advanced Photon Source data-management (DM) system and high-performance software to rapidly reduce area-detector data to multi-tau and two-time correlation functions in near real time, providing human-in-the-loop feedback to experimenters. The utility and performance of the workflow system are demonstrated via its application to a variety of small-angle XPCS measurements acquired from different detectors in different XPCS measurement modalities. The XSPA-500k detector, the software and the DM workflow system allow for the efficient acquisition and reduction of up to ∼109 area-detector data frames per day, facilitating the application of XPCS to measuring samples with weak scattering and fast dynamics.

APS Data Management System.

As the capabilities of modern X-ray detectors and acquisition technologies increase, so do the data rates and volumes produced at synchrotron beamlines. This brings into focus a number of challenges related to managing data at such facilities, including data transfer, near real-time data processing, automated processing pipelines, data storage, handling metadata and remote user access to data. The Advanced Photon Source Data Management System software is designed to help beamlines deal with these issues. This paper presents the system architecture and describes its components and functionality; the system's current usage is discussed, examples of its use have been provided and future development plans are outlined.

Distributed X-ray photon correlation spectroscopy data reduction using Hadoop MapReduce.

Multi-speckle X-ray photon correlation spectroscopy (XPCS) is a powerful technique for characterizing the dynamic nature of complex materials over a range of time scales. XPCS has been successfully applied to study a wide range of systems. Recent developments in higher-frame-rate detectors, while aiding in the study of faster dynamical processes, creates large amounts of data that require parallel computational techniques to process in near real-time. Here, an implementation of the multi-tau and two-time autocorrelation algorithms using the Hadoop MapReduce framework for distributed computing is presented. The system scales well with regard to the increase in the data size, and has been serving the users of beamline 8-ID-I at the Advanced Photon Source for near real-time autocorrelations for the past five years.

Scientific data exchange: a schema for HDF5-based storage of raw and analyzed data.

Data Exchange is a simple data model designed to interface, or `exchange', data among different instruments, and to enable sharing of data analysis tools. Data Exchange focuses on technique rather than instrument descriptions, and on provenance tracking of analysis steps and results. In this paper the successful application of the Data Exchange model to a variety of X-ray techniques, including tomography, fluorescence spectroscopy, fluorescence tomography and photon correlation spectroscopy, is described.

Imaging of mitochondrial and non-mitochondrial responses in cultured rat hippocampal neurons exposed to micromolar concentrations of TMRM.

Tetramethylrhodamine methyl ester (TMRM) is a fluorescent dye used to study mitochondrial function in living cells. Previously, we reported that TMRM effectively labeled mitochondria of neurons deep within mouse brain slices. Use of micromolar concentration of dye, which was required to get sufficient staining for two-photon imaging, resulted in typical fluctuations of TMRM. With prolonged exposure, we recorded additional responses in some neurons that included slow oscillations and propagating waves of fluorescence. (Note: We use the terms "fluctuation" to refer to a change in the fluorescent state of an individual mitochondrion, "oscillation" to refer to a localized change in fluorescence in the cytosol, and "wave" to refer to a change in cytosolic fluorescence that propagated within a cell. Use of these terms does not imply any underlying periodicity.) In this report we describe similar results using cultured rat hippocampal neurons. Prolonged exposure of cultures to 2.5 µM TMRM produced a spontaneous increase in fluorescence in some neurons, but not glial cells, after 45-60 minutes that was followed by slow oscillations, waves, and eventually apoptosis. Spontaneous increases in fluorescence were insensitive to high concentrations of FCCP (100 µM) and thapsigargin (10 µM) indicating that they originated, at least in part, from regions outside of mitochondria. The oscillations did not correlate with changes in intracellular Ca(2+), but did correlate with differences in fluorescence lifetime of the dye. Fluorescence lifetime and one-photon ratiometric imaging of TMRM suggested that the spontaneous increase and subsequent oscillations were due to movement of dye between quenched (hydrophobic) and unquenched (hydrophilic) compartments. We propose that these movements may be correlates of intracellular events involved in early stages of apoptosis.

Mild hypothermia during hemorrhagic shock in rats improves survival without significant effects on inflammatory responses.

OBJECTIVE: To explore the hypothesis that the survival benefit of mild, therapeutic hypothermia during hemorrhagic shock is associated with inhibition of lipid peroxidation and the acute inflammatory response. DESIGN: Prospective and randomized. SETTING: Animal research facility. SUBJECTS: Male Sprague-Dawley rats. INTERVENTIONS: Rats underwent pressure-controlled (mean arterial pressure 40 mm Hg) hemorrhagic shock for 90 mins. They were randomized to normothermia (38.0 +/- 0.5 degrees C) or mild hypothermia (33-34 degrees C from hemorrhagic shock 20 mins to resuscitation time 12 hrs). Rats were killed at resuscitation time 3 or 24 hrs. MEASUREMENTS AND MAIN RESULTS: All seven rats in the hypothermia group and seven of 15 rats in the normothermia group survived to 24 hrs (p <.05). Hypothermic rats had lower serum potassium and higher blood glucose concentrations at 90 mins of hemorrhagic shock (p <.05). At resuscitation time 24 hrs, the hypothermia group had less liver injury (based on serum concentrations of ornithine carbamolytransferase and liver histology) and higher blood glucose than the normothermia group (p <.05). There were no differences in serum free 8-isoprostane (a marker of lipid peroxidation by free radicals) between the two groups at either baseline or resuscitation time 1 hr. Serum concentrations of interleukin- 1 beta, interleukin-6, and tumor necrosis factor-alpha peaked at resuscitation time 1 hr. Tumor necrosis factor-alpha concentrations were higher (p <.05) at resuscitation time 1 hr in the hypothermia group compared with the normothermic group. Serum cytokine concentrations were not different between survivors and nonsurvivors in the normothermia group. Serum cytokine concentrations returned to baseline values in both groups by 24 hrs. There were no differences in the number of neutrophils in the lungs or the small intestine between the groups. More neutrophils were found in the lungs at resuscitation time 3 hrs than at resuscitation time 24 hrs in both groups (p <.01). CONCLUSIONS: These data suggest that lipid peroxidation and systemic inflammatory responses to hemorrhagic shock are minimally influenced by mild hypothermia, although liver injury is mitigated and survival improved. Other mechanisms of benefit from mild hypothermia need to be explored.