Artificial Structure Research Articles

Application of Hyperspectral Image for Monitoring in Coastal Area with Deep Learning: A Case Study of Green Algae on Artificial Structure

Remote sensing is a powerful technique for classifying and quantifying objects. However, the elaborate classification of objects in coastal waters with complex structures is still challenging due to the high possibility of class mixing. The classification through the hyperspectral images can be a reasonable alternative to problems related to such precise classification work because it has high spectral resolution over a wide bandwidth. This study introduced the results of the case study using a novel method to classify green algae on an artificial structure based on hyperspectral data and deep-learning models. The spectral characteristics of the attached green algae on the artificial structure were observed using a ground-based hyperspectral camera. The observed image had a total of three classes (concrete, dense green algae, and sparse green algae). A certain area of the image was used as learning data to create classification models for three classes. The classification models were created from one machine-learning (support vector machine, SVM) and two deep-learning models (convolutional neural network, CNN; and dense convolutional network, DenseNet). As a result, the performance for the classification results of green algae predicted from two deep-learning models was higher than that of the machine-learning model. Additionally, the deep-learning model successfully classified the interface area between concrete and green algae. This study suggests that the combination of hyperspectral data and deep learning could enable more precise classification of objects in coastal areas.

In silico formulation optimization and particle engineering of pharmaceutical products using a generative artificial intelligence structure synthesis method

Pharmaceutical drug dosage forms are critical for ensuring the effective and safe delivery of active pharmaceutical ingredients to patients. However, traditional formulation development often relies on extensive lab and animal experimentation, which can be time-consuming and costly. This manuscript presents a generative artificial intelligence method that creates digital versions of drug products from images of exemplar products. This approach employs an image generator guided by critical quality attributes, such as particle size and drug loading, to create realistic digital product variations that can be analyzed and optimized digitally. This paper shows how this method was validated through two case studies: one for the determination of the amount of material that will create a percolating network in an oral tablet product and another for the optimization of drug distribution in a long-acting HIV inhibitor implant. The results demonstrate that the generative AI method accurately predicts a percolation threshold of 4.2% weight of microcrystalline cellulose and generates implant formulations with controlled drug loading and particle size distributions. Comparisons with real samples reveal that the synthesized structures exhibit comparable particle size distributions and transport properties in release media.

BiLSTM-Filt: Neural network for radar word segmentation

Radar word extraction is the analysis foundation for multi-function radars (MFRs) in electronic intelligence (ELINT). Although neural networks enhance performance in radar word extraction, current research still faces challenges from complex electromagnetic environments and unknown radar words. Therefore, in this paper, we propose a promising two-stage radar word extraction framework, consisting of segmentation and recognition. To fill the vacancy of radar word segmentation, we establish the mathematical model from the time series analysis viewpoint and design a novel segmentation neural network based on Bi-direction Long Short-Term Memory with a filter module (BiLSTM-Filt). Specific radar word structure characteristics are extracted by training the network and applied for detecting radar words in the pulse train. To further improve segmentation performance, a bounding box regression method is designed to merge information from sub-region structures. Simulation experiments on a typical MFR, Mercury, reveal that the proposed method can outperform the baseline methods within complex electromagnetic environments, containing corrupted environments, various pulse backgrounds, and variable pulse train lengths. Due to the artificial design structure, the proposed method can also make a trial on unknown radar word segmentation.

Monitoring the Impact of Artificial Structure on Hydrogeological Environment: A Case Study of Hydraulic Tunnel at Pirot Hydropower Plant.

Artificial objects, particularly tunnels used for water transport under pressure, impact the geological and hydrogeological environment to a greater or lesser extent, and it is vital to assess their contributions to groundwater quality. Although tunnels are typically lined with concrete, their interaction with the hydrogeological environment intensifies over time. In this study, the detailed spatiotemporal monitoring of all hydrogeological features within the potential influence zone of the hydraulic tunnel of the Pirot Hydropower Plant has been conducted in order to determine the degree of interaction between the artificial object and the natural environment in real time, and to assess the correlation between monitored parameters. Natural conditions of the environment were defined, as well as potential changes through the observing groundwater regimes. The monitoring network included observations of groundwater regimes at seven springs located in close proximity to the hydraulic tunnel, within the tunnel, at three piezometers, and along the river, while methods employed were hydrological monitoring, physicochemical monitoring, and groundwater piezometer sensing. Cross-correlation analysis has been applied for assessing the impact of precipitation dynamics on the spring discharge regime. The results indicate a direct influence of the tunnel on the hydrogeological environment, proving the consistency and high correlation between the monitored parameters.

Atomistic simulations of polysaccharide materials for insights into their crystal structure, nanostructure, and dissolution mechanism

AbstractCrystalline polysaccharides are abundant in nature and can be transformed into highly functional materials. However, the molecular basis for the formation of higher-order structures remains unclear. Computer simulation is an advanced tool for modeling macromolecular structures, and the atomistic simulations provide valuable information on the crystalline polysaccharides. Fiber deformation, crystalline transition, and novel nanostructures of cellulose were characterized through molecular dynamics simulations and density functional theory calculations of models of molecular chain sheets extracted from the crystal structure of the cellulose polymorphs. Extended ensemble molecular dynamics simulations were applied to analyze the artificial crystal structure of non-natural amylose analog polysaccharides, revealing the hexagonal packing of double helices through the self-assembly of molecular chains dispersed in aqueous solution. Dissolution simulations of the cellulose and chitin crystalline fibers revealed that the anions of ionic liquids, with their solvation power, played a key role in the cleavage of intermolecular hydrogen bonds in the crystal structure, whereas the cations contributed to irreversible molecular chain dispersion. The good correlation between the actual solubility of polysaccharides and the predicted number of intermolecular hydrogen bonds prompted the development of a platform that combined simulations and machine learning for high-throughput screening of solvents for cellulose and chitin.

Au Nanorings/TiO2 NPs@MXene-Based Metasurfaces with a Magnetic Mirror-Modulated ECL Strategy for Extracellular Vesicle Detection.

A metasurface as an artificial electromagnetic structure can concentrate optical energy into nanometric volumes to strongly enhance the light-matter interaction, which has been becoming a powerful platform for optical sensing, nonlinear effects, and quantum optics. Herein, we developed a novel hybrid plasmonic-dielectric metasurface consisting of Au nanorings (Au NRs) and TiO2 nanoparticles derived from MXene (TiO2 NPs@MXene). The hybrid metasurface simultaneously benefited from the high near-field enhancement effect of plasmonic materials and the low loss of dielectric materials. Furthermore, the optical modulation efficiency of the hybrid metasurface can be regulated by a magnetic mirror configuration. The magnetic mirror acted like a mirror, confining the electrons to a limited region and increasing the density of the surface plasmon. Moreover, the electrochemiluminescence (ECL) of the Cu2BDC metal-organic framework (Cu2BDC-MOF) served as a light source for the Au NRs/TiO2 NPs@MXene metasurface. Due to the exceptional light manipulation capability of the hybrid metasurface and the coordination of the magnetic mirror, the isotropic ECL signal can be dynamically amplified and converted into polarized emission. Finally, a metasurface-regulated ECL (MECL)-based biosensor with a dual-positive membrane protein recognition strategy was developed for the accurate identification of gastric cancer-derived extracellular vesicles. The novel MECL research opened up a new route in the realization of dynamically tunable metasurfaces for optical sensing and novel nanophotonic devices.

Inverse design of metamaterial for dual-band infrared camouflage along with dual-band thermal management

The optical metamaterial for infrared camouflage with thermal management is an artificial structure available to not only evade the detection of various thermal imagers operating in distinct infrared wavebands but also ensure the thermal stability of the target. However, designing such an optical metamaterial is challenging due to the necessity to optimize numerous structural parameters simultaneously, especially metamaterials that are compatible with multiple functions. Herein, this paper proposes a novel inverse design method, combining African vultures optimization algorithm (AVOA) with rigorous coupled-wave analysis (RCWA), to design an optical metamaterial capable of achieving infrared camouflage and thermal management compatibly. The proposed metamaterial enables dual-band infrared camouflage for mid-wave infrared (MWIR, 3–5 μm) and long-wave infrared (LWIR, 8–14 μm), along with thermal management through dual-band non-atmospheric windows of 5–8 μm and 14–17 μm. Comparing to other intelligent optimization algorithms, the AVOA is exceptionally efficient, which only takes less time than others to get better results. This work not only provides theoretical guidance to design an optical metamaterial for infrared camouflage with thermal management, but also has the potential to solve the multi-objective optimization problems of multi-band or wide-band radiative modulation.

Feature Selection and Dimensionality Reduction for Large Data set Clustering using Self-Organizing Maps

Data mining is a form of analyzing the data which relates the newline techniques from fields like statistics machine learning databases artificial newline intelligence etc Clustering is one of the most important data mining newline techniques in which cluster of objects are grouped based on their similarity newline "Clustering is the process of accumulating the data records into considerable newline subclasses clusters in a way which enhances the relationship within clusters newline and reduces the similarity among two different clusters. The activity occurs every new academic year and schools with plenty of new students registered may feel a bit overwhelmed with this grouping assignment. A decision support system which can automatically perform grouping on a list of students may be able to help the school’s staffs with this repetitive task. A self-organizing map (SOM) is an example of unsupervised learning algorithm using an artificial neural network structure to produce a low dimensional representation from a given input.

AlphaFold2 Predicts Alternative Conformation Populations in Green Fluorescent Protein Variants.

Artificial intelligence-based protein structure prediction methods such as AlphaFold2 have emerged as powerful tools for characterizing proteins sequence-structure relationship offering unprecedented opportunities for the molecular interpretation of biological and biochemical phenomena. While initially confined to providing a static representation of proteins through their global free-energy minimum, AlphaFold2 has demonstrated the ability to partially sample conformational landscapes, providing insights into protein dynamics, which is fundamental for interpreting and potentially tuning the function of natural and artificial proteins. In this study, we show that targeted column masking of AlphaFold2's multiple sequence alignment enables the characterization and estimation of the population ratio of the two main conformations of engineered green fluorescent proteins with alternative β-strands. The possibility of quickly estimating relative populations through AlphaFold2 predictions is expected to speed-up the computational design of related systems for sensing applications.

Design of a optimal robust adaptive neural network-based fractional-order PID controller for H-bridge single-phase inverter

Adaptive Neural Network- based Fractional-order PID Control (NN-FO-PID) approach is designed for H-bridge inverter. This inverter has an LC filter to decrease the level of Total Harmonic Distortion (THD) that can affect the efficiency of the system. In addition, the reduction of THD and stability insurance of the filter are challenging performances. To fulfill this function, a fractional order proportional integrated derivative (FO-PID) controller is developed for the inverter. A few merits of the Fractional-Order notion as a useful technique include reduced sensitivity to noise and parametric fluctuation; however, for a wider range of disturbances, such as noise, this approach shows an unsuitable practical application based on its fixed gain values. Moreover, having parameters uncertainties including parametric variations, load uncertainty, supply voltage variation uplifts this challenging condition severely, and the parameters need to be adjusted once more for more dependable operations. As a result, the control parameters must be optimized once more to provide ideal operations. Here, an adaptive mechanism is proposed based on neural network structure to optimize the gains of the FO-PID controller for better performances. In real applications, this approach has some benefits, since it uses the Black-box technique, which does not necessitate a precise mathematical model of the system, resulting in a reduced computational burden, simple implementation, and reduced dependence on the model’s states. Even under extremely difficult circumstances, the artificial neural network structure effectively optimizes the FO-PID gains in real-time. This benefit can reduce the amount of THD-level for the inverter, properly. Additionally, to have a proper response in the first step condition, and trying to reduce the level of dangerous conditions, based on benefits of the ant lion optimization method, it is considered to select the initial values of the FO-PID controller gains. Here to verify the superiority of the proposed controller, PID and FO-PID controllers are offered to drive a comparison with this method, which are optimized using the PSO optimization algorithm. The analysis of simulation data shows that the suggested control strategy is appropriate for maintaining stability as well as adequately compensating for disturbances and uncertainties in the inverter. Furthermore, with the help of this controller, THD level decreased lower than 2 percent.

PhAI: A deep-learning approach to solve the crystallographic phase problem.

X-ray crystallography provides a distinctive view on the three-dimensional structure of crystals. To reconstruct the electron density map, the complex structure factors [Formula: see text] of a sufficiently large number of diffracted reflections must be known. In a conventional experiment, only the amplitudes [Formula: see text] are obtained, and the phases ϕ are lost. This is the crystallographic phase problem. In this work, we show that a neural network, trained on millions of artificial structure data, can solve the phase problem at a resolution of only 2 angstroms, using only 10 to 20% of the data needed for direct methods. The network works in common space groups and for modest unit-cell dimensions and suggests that neural networks could be used to solve the phase problem in the general case for weakly scattering crystals.

Artificial Trabecular Meshwork Structure Combining Melt Electrowriting and Solution Electrospinning.

The human trabecular meshwork (HTM) is responsible for regulating intraocular pressure (IOP) by means of gradient porosity. Changes in its physical properties, like increases in stiffness or alterations in the extracellular matrix (ECM), are associated with increases in the IOP, which is the primary cause of glaucoma. The complexity of its structure limits the engineered models to one-layered and simple approaches, which do not accurately replicate the biological and physiological cues related to glaucoma. Here, a combination of melt electrowriting (MEW) and solution electrospinning (SE) is explored as a biofabrication technique used to produce a gradient porous scaffold that mimics the multi-layered structure of the native HTM. Polycaprolactone (PCL) constructs with a height of 20-710 µm and fiber diameters of 0.7-37.5 µm were fabricated. After mechanical characterization, primary human trabecular meshwork cells (HTMCs) were seeded over the scaffolds within the subsequent 14-21 days. In order to validate the system's responsiveness, cells were treated with dexamethasone (Dex) and the rho inhibitor Netarsudil (Net). Scanning electron microscopy and immunochemistry staining were performed to evaluate the expected morphological changes caused by the drugs. Cells in the engineered membranes exhibited an HTMC-like morphology and a correct drug response. Although this work demonstrates the utility of combining MEW and SE in reconstructing complex morphological features like the HTM, new geometries and dimensions should be tested, and future works need to be directed towards perfusion studies.

African Swine Fever Virus Protein-Protein Interaction Prediction.

The African swine fever virus (ASFV) is an often deadly disease in swine and poses a threat to swine livestock and swine producers. With its complex genome containing more than 150 coding regions, developing effective vaccines for this virus remains a challenge due to a lack of basic knowledge about viral protein function and protein-protein interactions between viral proteins and between viral and host proteins. In this work, we identified ASFV-ASFV protein-protein interactions (PPIs) using artificial intelligence-powered protein structure prediction tools. We benchmarked our PPI identification workflow on the Vaccinia virus, a widely studied nucleocytoplasmic large DNA virus, and found that it could identify gold-standard PPIs that have been validated in vitro in a genome-wide computational screening. We applied this workflow to more than 18,000 pairwise combinations of ASFV proteins and were able to identify seventeen novel PPIs, many of which have corroborating experimental or bioinformatic evidence for their protein-protein interactions, further validating their relevance. Two protein-protein interactions, I267L and I8L, I267L__I8L, and B175L and DP79L, B175L__DP79L, are novel PPIs involving viral proteins known to modulate host immune response.

An improved artificial spin ice structure for restoring ice degeneracy

It is known that rare-earth-based pyrochlore oxide may accommodate the well-defined two-in-two-out spin ice state with its tetrahedral unit. Low-energy excitation is argued to favor the highly concerned monopole state which attracts essential attention. However, such an excitation cannot be directly tracked and imaged using advanced characterizations, raising challenges to our understanding of the physics of monopoles. In this work, we propose an improved two-dimensional artificial spin ice structure on the Shastry–Sutherland lattice to restore the degeneracy of realistic pyrochlore systems. Such a structure avoids the deficiency of inequivalent nearest and next-nearest exchanges in the planar quadrate unit, which, however, is equivalent to the tetrahedral unit of realistic pyrochlore oxides. Therefore, this spin ice model restores state degeneracy that is lost in conventional planar artificial spin ice structures, representing an improved simulator of real spin ice systems. Our careful investigations of such improved structures reveal the rich physics of spin ice excitations, including the phase diagram, which allows different ordered phases and interesting critical phase transitions between spin ice phase I and phase II. Energy spectrum analysis suggests that restoration of state degeneracy substantially reduces monopole excitation energy, resulting in a striking monopole emergency at the critical point. Furthermore, the emergent spin dimer phase in this improved model allows high-density monopole excitations and exhibits high-correlated monopole fluid states.

Artificial antibody-antigen-directed immobilization of α-amylase to hydrolyze starch for cascade reduction of 2-nitro-4-methylphenol to 2-amino-4-methylphenol

Nitrophenol is a hazardous substance that poses a threat to the environment and human health, and its treatment has attracted widespread attention. The purpose of this study is to establish an environmentally friendly α-amylase system for the hydrolysis of starch to reduce nitrophenol to aminophenol through cascade reactions. The α-amylase system was obtained through artificial antibody-antigen-directed immobilization, including the synthesis of artificial antibodies, synthesis of artificial antigens, and affinity assembly. In this process, catechol and protocatechuic aldehyde were used to prepare artificial antibodies and artificial antigens respectively through polymerization and Schiff base reactions. Then, artificial antibodies captured the catechol in the artificial antigen structure to form immobilized α-amylases. Compared with free α-amylase, the immobilized α-amylase showed a good reusability and excellent regenerative ability. Subsequently, the immobilized α-amylase were used in the reaction of catalyzing starch hydrolysis to synthesize 2-amino-4-methylphenol, and the yield of 2-amino-4-methylphenol was 58.88 ± 0.19 %. After 5 consecutive catalytic reactions, a yield of 47.61 ± 1.27 % can still be achieved.

Acoustic spin-controlled orbital rotations in double spiral acoustic beams

Similar to optical spin-orbit interactions (SOIs), acoustic SOIs are anticipated to offer fresh perspectives and capabilities for acoustic manipulation beyond conventional scalar degrees of freedom. However, the acoustic extrinsic SOIs caused by particular properties of the medium were seldom explored. Here, the acoustic extrinsic SOI is observed in a double spiral acoustic beam (DSAB), as evidenced by the rotation of the spatial intensity pattern along the propagation axis. The interaction of the acoustic plane wave with the well-designed artificial flat structure generates two non-paraxial focused acoustic vortices (NFAVs) with different spin angular momentums. The coaxial coupling between them leads to acoustic spin-controlled orbital rotation (SOR). Theoretical formulations, supported by numerical simulations and experimental results, are provided to demonstrate the validity of acoustic SOR. Our work provides new perspectives and capabilities for understanding sound processing, and may open an avenue for the development of spin-orbit acoustics.

Increasing the lifetime of confined electronic states in an artificial quantum structure

Increasing the lifetime of confined electronic states in an artificial quantum structure

Physical structure of the environment contributes to the development of diversity of microalgal assemblages

Aquatic macrophytes form a three dimensional complex structure in the littoral zones of lakes, with many physical, chemical and biological gradients and interactions. This special habitat harbours a unique microalgal assemblage called metaphyton, that differs both from the phytoplankton of the pelagial and from the benthic assemblages whose elements are tightly attached to the substrates. Since metaphytic assemblages significantly contribute to the diversity of lakes’ phytoplankton, it is crucial to understand and disentangle those mechanisms that ensure their development. Therefore, we focused on the question of how a single solid physical structure contribute to maintaining metaphytic assemblages. Using a laboratory experiment we studied the floristic and functional differences of microalgal assemblages in microcosms that simulated the conditions that an open water, a complex natural macrophyte stand (Utricularia vulgaris L.), or an artificial substrate (cotton wool) provide for them. We inoculated the systems with a species rich (> 326 species) microalgal assemblage collected from a eutrophic oxbow lake, and studied the diversity, trait and functional group composition of the assemblages in a 24 day long experimental period. We found that both natural and artificial substrates ensured higher species richness than the open water environment. Functional richness in the open water environment was lower than in the aquaria containing natural macrophyte stand but higher than in which cotton wool was placed. This means that the artificial physical structure enhanced functional redundancy of the resident functional groups. Elongation measures of microalgal assemblages showed the highest variation in the microcosms that simulated the open water environment. Our results suggest that assembly of metaphytic algal communities is not a random process, instead a deterministic one driven by the niche characteristics of the complex three dimensional structure created by the stands of aquatic macrophytes.

Design of a Neural Network Model for Point-Defect Microcavities in Two-Dimensional Silicon-Based Dielectric Column Photonic Crystals

Currently, circuits are becoming more and more highly integrated, but current electronic chip technology has difficulty in meeting the requirements for increased data transmission speed and capacity because of its characteristics of increased energy loss due to interactions between electronic components. In contrast, photonic technology is a promising solution due to its characteristics of high speed, wide bandwidth, and low interaction. Photonic crystals are materials with an artificial periodic dielectric structure, with photonic bandgap and localization properties that are critical to their performance. Photonic crystals, which use photons as an information transfer medium, allow flexible control of photon propagation, just as electrons are controlled in semiconductors, and the study of multifunctional photonic crystals is important for the construction of integrated optical circuits. This paper proposes a neural network-based model to analyze and predict the optical properties of point defect microcavities in 2D silicon-based dielectric column photonic crystals. By modeling the complex relationship between the structural parameters of the photonic crystal and its optical response, a theoretical approach for the design of 2D silicon dielectric column photonic crystals can be provided.

Conserved cysteine residues in Kaposi's sarcoma herpesvirus ORF34 are necessary for viral production and viral pre-initiation complex formation.

Kaposi's sarcoma herpesvirus (KSHV) ORF34 plays a significant role as a component of the viral pre-initiation complex (vPIC), which is indispensable for late gene expression across beta and gamma herpesviruses. Although the key role of ORF34 within the vPIC and its function as a hub protein have been recognized, further clarification regarding its specific contribution to vPIC functionality and interactions with other components is required. This study employed a deep-learning algorithm-assisted structural model of ORF34, revealing highly conserved amino acid residues across human beta- and gamma-herpesviruses localized in structured domains. Thus, we engineered ORF34 alanine-scanning mutants by substituting conserved residues with alanine. These mutants were evaluated for their ability to interact with other vPIC factors and restore viral production in cells harboring the ORF34-deficient KSHV-BAC. Our experimental results highlight the crucial role of the 4 cysteine residues conserved in ORF34: a tetrahedral arrangement consisting of a pair of C-Xn-C consensus motifs. This suggests the potential incorporation of metal cations in interacting with ORF24 and ORF66 vPIC components, facilitating late gene transcription, and promoting overall virus production by capturing metal cations. In summary, our findings underline the essential role of conserved cysteines in KSHV ORF34 for effective vPIC assembly and viral replication, thereby enhancing our understanding of the complex interplay between the vPIC components.