Predicting Effective Diffusivity of Porous Media from Images by Deep Learning

Investigation of the interface structure and filler network formation of liquid silicone rubber/silica nanocomposites based on ATR-FTIR spectroscopy and chemometrics

Nano-silica's strong reinforcing effect on silicone rubber has led to its extensive use in composite synthesis. However, the mechanism of bound rubber formation at the interface of the two components, as well as the correlation between microstructure and mechanical properties, are poorly understood, and traditional characterization methods struggle to clarify these aspects. A chemometric method was applied to rapidly characterize silicone rubber/silica nanocomposites, and ATR-FTIR spectral analysis revealed that with increasing extraction time, the rubber component gradually decreased while the silica component remained stable and a new component emerged. Through analysis, it is hypothesized that this new component is the interfacial bonding part of the composite. This finding is significant as it deepens our understanding of the composite's structure-property relationships. It also offers a fresh approach for characterizing the interfacial bonding in silicone rubber/silica nanocomposites, potentially facilitating the development of high-performance materials.

Mechano-diffusion of particles in stretchable hydrogels

Precise control over particle diffusion is promising for diverse modern technologies. Traditionally, particle diffusion is governed by the inherent properties of a liquid medium, limiting versatility and controllability. Here, we report a mechano-diffusion mechanism that harnesses mechanical deformation to control particle diffusion in stretchable hydrogels with a significantly enlarged tuning ratio and a highly expanded tuning freedom. The working principle is to leverage the mechanical deformation of stretchable hydrogels for modulating the polymer network's geometric transformation and the polymer chain's energy modulation, which synergistically tunes the energy barrier for particle diffusion. Using a model particle-hydrogel material system and a customized mechano-diffusion characterization platform, we demonstrate that tension loads can enhance the diffusivity of gold nanoparticles up to 22 times, far exceeding that in traditional liquid medium and by external fields. Additionally, we show particle diffusion in hydrogels can be manipulated spatiotemporally by controlling the hydrogels' stress state and loading rate. To further push the limit of the mechano-diffusion, we use experiment, theory, and simulation to explore particle diffusion in biaxially stretched hydrogels, simultaneously expanding the mesh size and reducing the energy barrier. The enlarged tuning ratio and expanded tuning freedom enable a model-guided drug delivery system for pressure-controlled release of drug molecules. Understanding this spatiotemporal mechano-diffusion mechanism will provide insights pertinent to a broad range of biological and synthetic soft materials.

Accurate lithofacies identification in deep shale gas reservoirs via an optimized neural network recognition model, Qiongzhusi Formation, southern Sichuan

The Lower Cambrian Qiongzhusi Formation is crucial for exploring deep shale gas in Sichuan, however, challenges in accurately classifying shale lithofacies have hindered its commercialization. To address this, the deep shale reservoirs of the Qiongzhusi Formation were categorized into five lithofacies, five microfacies, and two-sedimentary models utilizing thin sections, scanning electron microscopy, X-ray diffraction (XRD), and petrophysical parameters. Subsequently, various lithofacies identification methods for deep shale gas reservoirs were developed. The recognition performance of triangle and three-dimensional spatial distribution chart methods is poor. The recognition effects of neural network clustering analysis (the testing and validation datasets) are less than 80%, and the training dataset is only 82.6%. On the basis of the trigonometric features, three-dimensional spatial distribution features, and neural network clustering features of the dataset, an optimized neural network lithofacies recognition model was developed. The recognition accuracy of the testing, validation, and training datasets of the ONN model based on the DL principle yielded is greater than 80%. The model achieves a recognition accuracy (training dataset) of 89.9%, with an 85% accuracy rate for blind well lithofacies recognition. This model offers valuable guidance for the exploration and development of deep shale gas in the research area, providing a substantial reference for lithofacies identification in deep shale gas reservoirs of other regions.

Preferential adsorption of noble gases in zeolitic tuff with variable saturation: A modeling study of counter-intuitive diffusive-adsorptive behavior

Noble gas transport through geologic media has important applications in the prediction and characterization of measured gas signatures related to underground nuclear explosions (UNEs). Retarding processes such as adsorption can cause significant species fractionation of radionuclide gases, which has implications for measured and predicted signatures used to distinguish radioxenon originating from civilian nuclear facilities or from UNEs. Accounting for the effects of variable water saturation in geologic media on tracer transport is one of the most challenging aspects of modeling gas transport because there is no unifying relationship for the associated tortuosity changes between different rock types, and reactive transport processes such as adsorption that are affected by the presence of water likewise behave differently between gas species. In this study, we perform numerical diffusive-adsorptive transport simulations to estimate gas transport parameters associated with bench-scale laboratory diffusion cell experiments measuring breakthrough in zeolitic and non-zeolitic rocks for a gaseous mixture of xenon, krypton, and SF6 at varying degrees of water saturation (Sw). Counter-intuitive transport behavior was observed in the zeolitic rock experiments whereby breakthrough concentrations were significantly higher when the core was partially saturated (Sw=17%) than under dry (Sw=0%) conditions. Breakthrough of xenon was especially retarded in the dry core - likely due to comparatively high affinity of xenon for zeolitic adsorption sites - and estimated effective diffusion coefficients for all gases were approximately an order of magnitude lower than what is predicted by porosity-tortuosity models. We propose the counter-intuitive behavior observed is because water infiltration of zeolite nanopores reduces both the adsorptive capacity of the rock and the tortuosity of connected flow paths. We developed a two-site competitive kinetic Langmuir adsorption reaction for the porous media transport simulator in order to constrain transport parameters within zeolitic tuff, where differential adsorption to zeolite and non-zeolite pores was observed. We determined that liquid saturation-dependent diffusive-adsorptive transport is affected by subtle and at times competing processes that are specific to different gases, which have a significant overall influence on effective transport parameters.

Physics-informed and data-driven discovery of governing equations for complex phenomena in heterogeneous media

Rapid evolution of sensor technology, advances in instrumentation, and progress in devising data-acquisition software and hardware are providing vast amounts of data for various complex phenomena that occur in heterogeneous media, ranging from those in atmospheric environment, to large-scale porous formations, and biological systems. The tremendous increase in the speed of scientific computing has also made it possible to emulate diverse multiscale and multiphysics phenomena that contain elements of stochasticity or heterogeneity, and to generate large volumes of numerical data for them. Thus, given a heterogeneous system with annealed or quenched disorder in which a complex phenomenon occurs, how should one analyze and model the system and phenomenon, explain the data, and make predictions for length and time scales much larger than those over which the data were collected? We divide such systems into three distinct classes. (i) Those for which the governing equations for the physical phenomena of interest, as well as data, are known, but solving the equations over large length scales and long times is very difficult. (ii) Those for which data are available, but the governing equations are only partially known, in the sense that they either contain various coefficients that must be evaluated based on the data, or that the number of degrees of freedom of the system is so large that deriving the complete equations is very difficult, if not impossible, as a result of which one must develop the governing equations with reduced dimensionality. (iii) In the third class are systems for which large amounts of data are available, but the governing equations for the phenomena of interest are not known. Several classes of physics-informed and data-driven approaches for analyzing and modeling of the three classes of systems have been emerging, which are based on machine learning, symbolic regression, the Koopman operator, the Mori-Zwanzig projection operator formulation, sparse identification of nonlinear dynamics, data assimilation combined with a neural network, and stochastic optimization and analysis. This perspective describes such methods and the latest developments in this highly important and rapidly expanding area and discusses possible future directions.

Thrombolytic and anticoagulant efficiencies of purified fibrinolytic enzyme produced from Cochliobolus hawaiiensis under solid-state fermentation

Cochliobolus hawaiiensis Alcorn Assiut University Mycological Centre 8606 was chosen from the screened 20 fungal species as the potent producer of fibrinolytic enzyme on skimmed-milk agar plates. The greatest enzyme yield was attained when the submerged fermentation (SmF) conditions were optimized, and it was around (39.7 U/mg protein). Moreover, upon optimization of fibrinolytic enzyme production under solid-state fermentation (SSF), the maximum productivity of fibrinolytic enzyme was greatly increased recorded a bout (405 U/mg protein) on sugarcane bagasse, incubation period of 5 days, moisture level of 100%, initial pH of salt basal medium 7.8, incubation temperature at 35°C, and supplementation of the salt basal medium with corn steep liquor (80％, v/v). The yield of fibrinolytic enzyme by C. hawaiiensis under SSF was higher than that of SmF with about 10.20-fold. The purification procedures of fibrinolytic enzyme by ammonium sulfate (70%), gel filtration, and ion-exchange columns chromatography caused a great increase in its specific activity to 2581.6 U/mg protein with an overall yield of 55.89%, 6.37 purification fold and molecular weight of 35 kDa. Maximal activity was recorded at pH 7 and 37°C. Significant pH stability was recorded at pH 6.6-7.2, and thermal stability was recorded at 33-41°C. The enzyme showed the highest affinity toward fibrin, with Vmax of 240 U/mL and an apparent Km value of 47.61 mmol. Mg2+ and Ca2+ moderately induced fibrinolytic activity, whereas Cu2+ and Zn2+ greatly suppressed the enzyme activity. The produced enzyme is categorized as serine protease and non-metalloprotease. The purified fibrinolytic enzyme showed efficient thrombolytic and antiplatelet aggregation activities by completely prevention and dissolution of the blood clot which confirmed by microscopic examination and amelioration of blood coagulation assays. These findings suggested that the produced fibrinolytic enzyme is a promising agent in management of blood coagulation disorders.

Deep learning for diffusion in porous media

We adopt convolutional neural networks (CNN) to predict the basic properties of the porous media. Two different media types are considered: one mimics the sand packings, and the other mimics the systems derived from the extracellular space of biological tissues. The Lattice Boltzmann Method is used to obtain the labeled data necessary for performing supervised learning. We distinguish two tasks. In the first, networks based on the analysis of the system's geometry predict porosity and effective diffusion coefficient. In the second, networks reconstruct the concentration map. In the first task, we propose two types of CNN models: the C-Net and the encoder part of the U-Net. Both networks are modified by adding a self-normalization module [Graczyk et al. in Sci Rep 12, 10583 (2022)]. The models predict with reasonable accuracy but only within the data type, they are trained on. For instance, the model trained on sand packings-like samples overshoots or undershoots for biological-like samples. In the second task, we propose the usage of the U-Net architecture. It accurately reconstructs the concentration fields. In contrast to the first task, the network trained on one data type works well for the other. For instance, the model trained on sand packings-like samples works perfectly on biological-like samples. Eventually, for both types of the data, we fit exponents in the Archie's law to find tortuosity that is used to describe the dependence of the effective diffusion on porosity.

A data-driven framework for permeability prediction of natural porous rocks via microstructural characterization and pore-scale simulation

Understanding the microstructure-property relationships of porous media is of great practical significance, based on which macroscopic physical properties can be directly derived from measurable microstructural informatics. However, establishing reliable microstructure-property mappings in an explicit manner is difficult, due to the intricacy, stochasticity, and heterogeneity of porous microstructures. In this paper, a data-driven computational framework is presented to investigate the inherent microstructure-permeability linkage for natural porous rocks, where multiple techniques are integrated together, including microscopy imaging, stochastic reconstruction, microstructural characterization, pore-scale simulation, feature selection, and data-driven modeling. A large number of 3D digital rocks with a wide porosity range are acquired from microscopy imaging and stochastic reconstruction techniques. A broad variety of morphological descriptors are used to quantitatively characterize pore microstructures from different perspectives, and they compose the raw feature pool for feature selection. High-fidelity lattice Boltzmann simulations are conducted to resolve fluid flow passing through porous media, from which reliable permeability references are obtained. The optimal feature set that best represents permeability is identified through a performance-oriented feature selection process, upon which a cost-effective surrogate model is rapidly fitted to approximate the microstructure-permeability mapping via data-driven modeling. This surrogate model exhibits great advantages over empirical/analytical formulas in terms of prediction accuracy and generalization capacity, which can predict reliable permeability values spanning four orders of magnitude. Besides, feature selection also greatly enhances the interpretability of the data-driven prediction model, from which new insights into the mechanism of how microstructural characteristics determine intrinsic permeability are obtained.

Coupled Lattice Boltzmann Modeling Framework for Pore-Scale Fluid Flow and Reactive Transport

In this paper, we propose a modeling framework for pore-scale fluid flow and reactive transport based on a coupled lattice Boltzmann model (LBM). We develop a modeling interface to integrate the LBM modeling code parallel lattice Boltzmann solver and the PHREEQC reaction solver using multiple flow and reaction cell mapping schemes. The major advantage of the proposed workflow is the high modeling flexibility obtained by coupling the geochemical model with the LBM fluid flow model. Consequently, the model is capable of executing one or more complex reactions within desired cells while preserving the high data communication efficiency between the two codes. Meanwhile, the developed mapping mechanism enables the flow, diffusion, and reactions in complex pore-scale geometries. We validate the coupled code in a series of benchmark numerical experiments, including 2D single-phase Poiseuille flow and diffusion, 2D reactive transport with calcite dissolution, as well as surface complexation reactions. The simulation results show good agreement with analytical solutions, experimental data, and multiple other simulation codes. In addition, we design an AI-based optimization workflow and implement it on the surface complexation model to enable increased capacity of the coupled modeling framework. Compared to the manual tuning results proposed in the literature, our workflow demonstrates fast and reliable model optimization results without incorporating pre-existing domain knowledge.

Inverse design of anisotropic spinodoid materials with prescribed diffusivity

The three-dimensional microstructure of functional materials determines its effective properties, like the mass transport properties of a porous material. Hence, it is desirable to be able to tune the properties by tuning the microstructure accordingly. In this work, we study a class of spinodoid i.e. spinodal decomposition-like structures with tunable anisotropy, based on Gaussian random fields. These are realistic yet computationally efficient models for bicontinuous porous materials. We use a convolutional neural network for predicting effective diffusivity in all three directions. We demonstrate that by incorporating the predictions of the neural network in an approximate Bayesian computation framework for inverse problems, we can in a computationally efficient manner design microstructures with prescribed diffusivity in all three directions.

Influence of the parameters of the convolutional neural network model in predicting the effective compressive modulus of porous structure

In recent years, the convolutional neural network (CNN) technique has emerged as an efficient new method for designing porous structure, but a CNN model generally contains a large number of parameters, each of which could influence the predictive ability of the CNN model. Furthermore, there is no consensus on the setting of each parameter in the CNN model. Therefore, the present study aimed to investigate the sensitivity of the parameters in the CNN model for the prediction of the mechanical property of porous structures. 10,500 samples of porous structure were randomly generated, and their effective compressive moduli obtained from finite element analysis were used as the ground truths to construct and train a CNN model. 8,000 of the samples were used to train the CNN model, 2000 samples were used for the cross-validation of the CNN model and the remaining 500 new structures, which did not participate in the CNN training process, were used to test the predictive power of the CNN model. The sensitivity of the number of convolutional layers, the number of convolution kernels, the number of pooling layers, the number of fully connected layers and the optimizer in the CNN model were then investigated. The results showed that the optimizer has the largest influence on the training speed, while the fully connected layer has the least impact on the training speed. Additionally, the pooling layer has the largest impact on the predictive ability while the optimizer has the least impact on the predictive ability. In conclusion, the parameters of the CNN model play an important role in the performance of the CNN model and the parameter sensitivity analysis can help optimize the CNN model to increase the computational efficiency.

Luminescent Lanthanide Probes for Inorganic and Organic Phosphates

Inorganic and organic phosphates-including orthophosphate, nucleotides, and DNA-are some of the most fundamental anions in cellular biology, regulating numerous processes of both medical and environmental significance. The characteristic long lifetimes of emitting lanthanides, including the brighter europium(III) and terbium(III), make them ideally suited for the development of molecular probes for the detection of phosphates directly in complex aqueous media. Moreover, given their high oxophilicity and the exquisite sensitivity of their quantum yields to their hydration number, those luminescent lanthanides are perfect for the detection of phosphates. Herein we discuss the principles that have guided the recent developments of molecular probes selective for inorganic or organic phosphates and how these lanthanide complexes facilitate the study of numerous biological processes.

Deep learning-based quasi-continuum theory for structure of confined fluids

Predicting the structural properties of water and simple fluids confined in nanometer scale pores and channels is essential in, for example, energy storage and biomolecular systems. Classical continuum theories fail to accurately capture the interfacial structure of fluids. In this work, we develop a deep learning-based quasi-continuum theory (DL-QT) to predict the concentration and potential profiles of a Lennard-Jones (LJ) fluid and water confined in a nano channel. The deep learning model is built based on a convolutional encoder-decoder network (CED) and is applied for high dimensional surrogate modeling to relate the fluid properties to the fluid-fluid potential. The CED model is then combined with the interatomic potential-based continuum theory to determine the concentration profiles of a confined LJ fluid and confined water. We show that the DL-QT model exhibits a robust predictive performance for a confined LJ fluid under various thermodynamic states and water confined in a nanochannel of different widths. The DL-QT model seamlessly connects the molecular physics at nanoscale with the continuum theory by using the deep learning model.

Construction of miRNA-lncRNA-mRNA co-expression network affecting EMT-mediated cisplatin resistance in ovarian cancer

Platinum resistance is one of the major concerns in ovarian cancer treatment. Recent evidence shows the critical role of epithelial-mesenchymal transition (EMT) in this resistance. Epithelial-like ovarian cancer cells show decreased sensitivity to cisplatin after cisplatin treatment. Our study prospected the association between epithelial phenotype and response to cisplatin in ovarian cancer. Microarray dataset GSE47856 was acquired from the GEO database. After identifying differentially expressed genes (DEGs) between epithelial-like and mesenchymal-like cells, the module identification analysis was performed using weighted gene co-expression network analysis (WGCNA). The gene ontology (GO) and pathway analyses of the most considerable modules were performed. The protein-protein interaction network was also constructed. The hub genes were specified using Cytoscape plugins MCODE and cytoHubba, followed by the survival analysis and data validation. Finally, the co-expression of miRNA-lncRNA-TF with the hub genes was reconstructed. The co-expression network analysis suggests 20 modules relating to the Epithelial phenotype. The antiquewhite4, brown and darkmagenta modules are the most significant non-preserved modules in the Epithelial phenotype and contain the most differentially expressed genes. GO, and KEGG pathway enrichment analyses on these modules divulge that these genes were primarily enriched in the focal adhesion, DNA replication pathways and stress response processes. ROC curve and overall survival rate analysis show that the co-expression pattern of the brown module's hub genes could be a potential prognostic biomarker for ovarian cancer cisplatin resistance.

Structure induced laminar vortices control anomalous dispersion in porous media

Natural porous systems, such as soil, membranes, and biological tissues comprise disordered structures characterized by dead-end pores connected to a network of percolating channels. The release and dispersion of particles, solutes, and microorganisms from such features is key for a broad range of environmental and medical applications including soil remediation, filtration and drug delivery. Yet, owing to the stagnant and opaque nature of these disordered systems, the role of microscopic structure and flow on the dispersion of particles and solutes remains poorly understood. Here, we use a microfluidic model system that features a pore structure characterized by distributed dead-ends to determine how particles are transported, retained and dispersed. We observe strong tailing of arrival time distributions at the outlet of the medium characterized by power-law decay with an exponent of 2/3. Using numerical simulations and an analytical model, we link this behavior to particles initially located within dead-end pores, and explain the tailing exponent with a hopping across and rolling along the streamlines of vortices within dead-end pores. We quantify such anomalous dispersal by a stochastic model that predicts the full evolution of arrival times. Our results demonstrate how microscopic flow structures can impact macroscopic particle transport.

A Lab on a Chip Experiment for Upscaling Diffusivity of Evolving Porous Media

Reactive transport modelling is a powerful tool to assess subsurface evolution in various energy-related applications. Upscaling, i.e., accounting for pore scale heterogeneities into larger scale analyses, remains one of the biggest challenges of reactive transport modelling. Pore scale simulations capturing the evolutions of the porous media over a wide range of Peclet and Damköhler number in combination with machine learning are foreseen as an efficient methodology for upscaling. However, the accuracy of these pore scale models needs to be tested against experiments. In this work, we developed a lab on a chip experiment with a novel micromodel design combined with operando confocal Raman spectroscopy, to monitor the evolution of porous media undergoing coupled mineral dissolution and precipitation processes due to diffusive reactive fluxes. The 3D-imaging of the porous media combined with pore scale modelling enabled the derivation of upscaled transport parameters. The chemical reaction tested involved the replacement of celestine by strontianite, whereby a net porosity increase is expected because of the smaller molar volume of strontianite. However, under our experimental conditions, the accessible porosity and consequently diffusivity decreased. We propose a transferability of the concepts behind the Verma and Pruess relationship to be applied to also describe changes of diffusivity for evolving porous media. Our results highlight the importance of calibrating pore scale models with quantitative experiments prior to simulations over a wide range of Peclet and Damköhler numbers of which results can be further used for the derivation of upscaled parameters.

Characterization of novel lncRNA muscle expression profiles associated with meat quality in beef cattle

The aim of this study was to identify novel lncRNA differentially expressed (DE) between divergent animals for beef tenderness and marbling traits in Nellore cattle. Longissimus thoracis muscle samples from the 20 most extreme bulls (of 80 bulls set) for tenderness, tender (n = 10) and tough (n = 10) groups, and marbling trait, high (n = 10) and low (n = 10) groups were used to perform transcriptomic analysis using RNA‐Sequencing. For tenderness, 29 lncRNA were DE (p‐value ≤ 0.01) in tough beef animals in relation to tender beef animals. We observed that genic lncRNAs, for example, lncRNA_595.1, were overlapping exonic part of the PICK gene, while lncRNA_3097.2 and lncRNA_3129.5 overlapped intronic part of the genes GADL1 and PSMD6. The lncRNA associated with PICK1, GADL1, and PMD6 genes were enriched in the pathways associated with the ionotropic glutamate receptor, gamma‐aminobutyric acid synthesis, and the ubiquitin–proteasome pathway. For marbling, 50 lncRNA were DE (p‐value ≤ 0.01) in high marbling group compared with low marbling animals. The genic lncRNAs, such as lncRNA_3191.1, were overlapped exonic part of the ITGAL gene, and the lncRNA_512.1, lncRNA_3721.1, and lncRNA_41.4 overlapped intronic parts of the KRAS and MASP1 genes. The KRAS and ITGAL genes were enriched in pathways associated with integrin signaling, which is involved in intracellular signals in response to the extracellular matrix, including cell form, mobility, and mediates progression through the cell cycle. In addition, the lincRNAs identified to marbling trait were associated with several genes related to calcium binding, muscle hypertrophy, skeletal muscle, lipase, and oxidative stress response pathways that seem to play a role important in the physiological processes related to meat quality. These findings bring new insights to better understand the biology mechanisms involved in the gene regulation of these traits, which will be valuable for a further investigation of the interactions between lncRNA and mRNAs, and of how these interactions may affect meat quality traits.

Machine Learning-Assisted Design of Material Properties

Designing functional materials requires a deep search through multidimensional spaces for system parameters that yield desirable material properties. For cases where conventional parameter sweeps or trial-and-error sampling are impractical, inverse methods that frame design as a constrained optimization problem present an attractive alternative. However, even efficient algorithms require time- and resource-intensive characterization of material properties many times during optimization, imposing a design bottleneck. Approaches that incorporate machine learning can help address this limitation and accelerate the discovery of materials with targeted properties. In this article, we review how to leverage machine learning to reduce dimensionality in order to effectively explore design space, accelerate property evaluation, and generate unconventional material structures with optimal properties. We also discuss promising future directions, including integration of machine learning into multiple stages of a design algorithm and interpretation of machine learning models to understand how design parameters relate to material properties. Expected final online publication date for the Annual Review of Chemical and Biomolecular Engineering, Volume 13 is October 2022. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.

Wastewater treatment by microalgae

The growth of the world's population increases the demand for fresh water, food, energy, and technology, which in turn leads to increasing amount of wastewater, produced both by domestic and industrial sources. These different wastewaters contain a wide variety of organic and inorganic compounds which can cause tremendous environmental problems if released untreated. Traditional treatment systems are usually expensive, energy demanding and are often still incapable of solving all challenges presented by the produced wastewaters. Microalgae are promising candidates for wastewater reclamation as they are capable of reducing the amount of nitrogen and phosphate as well as other toxic compounds including heavy metals or pharmaceuticals. Compared to the traditional systems, photosynthetic microalgae require less energy input since they use sunlight as their energy source, and at the same time lower the carbon footprint of the overall reclamation process. This mini-review focuses on recent advances in wastewater reclamation using microalgae. The most common microalgal strains used for this purpose are described as well as the challenges of using wastewater from different origins. We also describe the impact of climate with a particular focus on a Nordic climate.

Application of Machine Learning to Solid Particle Erosion of APS-TBC and EB-PVD TBC at Elevated Temperatures

Machine learning (ML) and deep learning (DL) for big data (BD) management are currently viable approaches that can significantly help in high-temperature materials design and development. ML-DL can accumulate knowledge by learning from existing data generated through multi-physics modelling (MPM) and experimental tests (ETs). DL mainly involves analyzing nonlinear correlations and high-dimensional datasets implemented through specifically designed numerical algorithms. DL also makes it possible to learn from new data and modify predictive models over time, identifying anomalies, signatures, and trends in machine performance, develop an understanding of patterns of behaviour, and estimate efficiencies in a machine. Machine learning was implemented to investigate the solid particle erosion of both APS (air plasma spray) and EB-PVD (electron beam physical vapour deposition) TBCs of hot section components. Several ML models and algorithms were used such as neural networks (NNs), gradient boosting regression (GBR), decision tree regression (DTR), and random forest regression (RFR). It was found that the test data are strongly associated with five key factors as identifiers. Following test data collection, the dataset is subjected to sorting, filtering, extracting, and exploratory analysis. The training and testing, and prediction results are analysed. The results suggest that neural networks using the BR model and GBR have better prediction capability.

Therapeutic potential of phytoconstituents of edible fruits in combating emerging viral infections

Plant-derived bioactive molecules display potential antiviral activity against various viral targets including mode of viral entry and its replication in host cells. Considering the challenges and search for antiviral agents, this review provides substantiated data on chemical constituents of edible fruits with promising antiviral activity. The bioactive constituents like naringenin, mangiferin, α-mangostin, geraniin, punicalagin, and lectins of edible fruits exhibit antiviral effect by inhibiting viral replication against IFV, DENV, polio, CHIKV, Zika, HIV, HSV, HBV, HCV, and SARS-CoV. The significance of edible fruit phytochemicals to block the virulence of various deadly viruses through their inhibitory action against the entry and replication of viral genetic makeup and proteins are discussed. In view of the antiviral property of active constituents of edible fruits which can strengthen the immune system and reduce oxidative stress, they are suggested to be diet supplements to combat various viral diseases including COVID-19. PRACTICAL APPLICATIONS: Considering the increasing threat of COVID-19, it is suggested to examine the therapeutic efficacy of existing antiviral molecules of edible fruits which may provide prophylactic and adjuvant therapy with their potential antioxidant, anti-inflammatory, and immune-modulatory effects. Several active molecules like geraniin, naringenin, (2R,4R)-1,2,4-trihydroxyheptadec-16-one, betacyanins, mangiferin, punicalagin, isomangiferin, procyanidin B2, quercetin, marmelide, jacalin lectin, banana lectin, and α-mangostin isolated from various edible fruits have showed promising antiviral properties against different pathogenic viruses. Especially flavonoid compounds extracted from edible fruits possess potential antiviral activity against a wide array of viruses like HIV-1, HSV-1 and 2, HCV, INF, dengue, yellow fever, NSV, and Zika virus infection. Hence taking such fruits or edible fruits and their constituents/compounds as dietary supplements could deliver adequate plasma levels in the body to optimize the cell and tissue levels and could lead to possible benefits for the preventive measures for this pandemic COVID-19 situation.

Design and Challenges of Sonodynamic Therapy System for Cancer Theranostics: From Equipment to Sensitizers

As a novel noninvasive therapeutic modality combining low-intensity ultrasound and sonosensitizers, sonodynamic therapy (SDT) is promising for clinical translation due to its high tissue-penetrating capability to treat deeper lesions intractable by photodynamic therapy (PDT), which suffers from the major limitation of low tissue penetration depth of light. The effectiveness and feasibility of SDT are regarded to rely on not only the development of stable and flexible SDT apparatus, but also the screening of sonosensitizers with good specificity and safety. To give an outlook of the development of SDT equipment, the key technologies are discussed according to five aspects including ultrasonic dose settings, sonosensitizer screening, tumor positioning, temperature monitoring, and reactive oxygen species (ROS) detection. In addition, some state-of-the-art SDT multifunctional equipment integrating diagnosis and treatment for accurate SDT are introduced. Further, an overview of the development of sonosensitizers is provided from small molecular sensitizers to nano/microenhanced sensitizers. Several types of nanomaterial-augmented SDT are in discussion, including porphyrin-based nanomaterials, porphyrin-like nanomaterials, inorganic nanomaterials, and organic-inorganic hybrid nanomaterials with different strategies to improve SDT therapeutic efficacy. There is no doubt that the rapid development and clinical translation of sonodynamic therapy will be promoted by advanced equipment, smart nanomaterial-based sonosensitizer, and multidisciplinary collaboration.

Prediction and Optimization of Ion Transport Characteristics in Nanoparticle-Based Electrolytes Using Convolutional Neural Networks

We develop a convolutional neural network (CNN) model to predict the diffusivity of cations in nanoparticle-based electrolytes and use it to identify the characteristics of morphologies that exhibit optimal transport properties. The ground truth data are obtained from kinetic Monte Carlo (kMC) simulations of cation transport parametrized using a multiscale modeling strategy. We implement deep learning approaches to quantitatively link the diffusivity of cations to the spatial arrangement of the nanoparticles. We then integrate the trained CNN model with a topology optimization algorithm for accelerated discovery of nanoparticle morphologies that exhibit optimal cation diffusivities at a specified nanoparticle loading, and we investigate the ability of the CNN model to quantitatively account for the influence of interparticle spatial correlations on cation diffusivity. Finally, by using data-driven approaches, we explore how simple descriptors of nanoparticle morphology correlate with cation diffusivity, thus providing a physical rationale for the observed optimal microstructures. The results of this study highlight the capability of CNNs to serve as surrogate models for structure-property relationships in composites with monodisperse spherical particles, which can in turn be used with inverse methods to discover morphologies that produce optimal target properties.

Characterising soft matter using machine learning

Machine learning is making a major impact in materials research. I review current progress across a selection of areas of ubiquitous soft matter. When applied to particle tracking, machine learning using convolution neural networks is providing impressive performance but there remain some significant problems to solve. Characterising ordered arrangements of particles is a huge challenge and machine learning has been deployed to create the description, perform the classification and tease out an interpretation using a wide array of techniques often with good success. In glass research, machine learning has proved decisive in quantifying very subtle correlations between the local structure around a site and the susceptibility towards a rearrangement event at that site. There are also beginning to be some impressive attempts to deploy machine learning in the design of composite soft materials. The discovery aspect of this new materials design meets the current interest in teaching algorithms to learn to extrapolate beyond the training data.

Predicting porosity, permeability, and tortuosity of porous media from images by deep learning

Convolutional neural networks (CNN) are utilized to encode the relation between initial configurations of obstacles and three fundamental quantities in porous media: porosity ([Formula: see text]), permeability (k), and tortuosity (T). The two-dimensional systems with obstacles are considered. The fluid flow through a porous medium is simulated with the lattice Boltzmann method. The analysis has been performed for the systems with [Formula: see text] which covers five orders of magnitude a span for permeability [Formula: see text] and tortuosity [Formula: see text]. It is shown that the CNNs can be used to predict the porosity, permeability, and tortuosity with good accuracy. With the usage of the CNN models, the relation between T and [Formula: see text] has been obtained and compared with the empirical estimate.

Consumption of reused vegetable oil intensifies BRCA1 mutations

Breast cancer (BC) is a foremost type of cancer in women globally with an increased mortality rate in developing countries. Information regarding hereditary factors, lifestyle, work environment, food habits, and personal history could be useful in diagnosing breast cancer. Among such food habits, the reuse of edible oil for preparing food is a common practice in any developing country. The repeated heating of oils enhances the oxidative degradation of oil to produce polyaromatic hydrocarbons (PAH) which could disrupt the redox balance and generate reactive oxygen species. These reactive toxic intermediates can lead to BRCA1 mutations that are responsible for breast cancer. Mutations in DNA are the main cause for the conversion of proto-oncogenes into oncogenes which leads to change in expression and an increase in cell proliferation wherein a normal cell gets transformed into a malignant neoplastic cell. This review summarizes the possible mechanism involved in the induction of breast cancer due to repeated heating of edible.

Predicting permeability via statistical learning on higher-order microstructural information

Quantitative structure-property relationships are crucial for the understanding and prediction of the physical properties of complex materials. For fluid flow in porous materials, characterizing the geometry of the pore microstructure facilitates prediction of permeability, a key property that has been extensively studied in material science, geophysics and chemical engineering. In this work, we study the predictability of different structural descriptors via both linear regressions and neural networks. A large data set of 30,000 virtual, porous microstructures of different types, including both granular and continuous solid phases, is created for this end. We compute permeabilities of these structures using the lattice Boltzmann method, and characterize the pore space geometry using one-point correlation functions (porosity, specific surface), two-point surface-surface, surface-void, and void-void correlation functions, as well as the geodesic tortuosity as an implicit descriptor. Then, we study the prediction of the permeability using different combinations of these descriptors. We obtain significant improvements of performance when compared to a Kozeny-Carman regression with only lowest-order descriptors (porosity and specific surface). We find that combining all three two-point correlation functions and tortuosity provides the best prediction of permeability, with the void-void correlation function being the most informative individual descriptor. Moreover, the combination of porosity, specific surface, and geodesic tortuosity provides very good predictive performance. This shows that higher-order correlation functions are extremely useful for forming a general model for predicting physical properties of complex materials. Additionally, our results suggest that artificial neural networks are superior to the more conventional regression methods for establishing quantitative structure-property relationships. We make the data and code used publicly available to facilitate further development of permeability prediction methods.

Phase transitions, percolation, fracture of materials, and deep learning

Percolation and fracture propagation in disordered solids represent two important problems in science and engineering that are characterized by phase transitions: loss of macroscopic connectivity at the percolation threshold p_{c} and formation of a macroscopic fracture network at the incipient fracture point (IFP). Percolation also represents the fracture problem in the limit of very strong disorder. An important unsolved problem is accurate prediction of physical properties of systems undergoing such transitions, given limited data far from the transition point. There is currently no theoretical method that can use limited data for a region far from a transition point p_{c} or the IFP and predict the physical properties all the way to that point, including their location. We present a deep neural network (DNN) for predicting such properties of two- and three-dimensional systems and in particular their percolation probability, the threshold p_{c}, the elastic moduli, and the universal Poisson ratio at p_{c}. All the predictions are in excellent agreement with the data. In particular, the DNN predicts correctly p_{c}, even though the training data were for the state of the systems far from p_{c}. This opens up the possibility of using the DNN for predicting physical properties of many types of disordered materials that undergo phase transformation, for which limited data are available for only far from the transition point.