Leveraging long short-term memory (LSTM)-based neural networks for modeling structure-property relationships of metamaterials from electromagnetic responses.
Ontology highlight
ABSTRACT:
SUBMITTER: Pillai P
PROVIDER: S-EPMC8452651 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:BACKGROUND:Recent studies have proposed deep learning techniques, namely recurrent neural networks, to improve biomedical text mining tasks. However, these techniques rarely take advantage of existing domain-specific resources, such as ontologies. In Life and Health Sciences there is a vast and valuable set of such resources publicly available, which are continuously being updated. Biomedical ontologies are nowadays a mainstream approach to formalize existing knowledge about entities, such as genes, chemicals, phenotypes, and disorders. These resources contain supplementary information that may not be yet encoded in training data, particularly in domains with limited labeled data. RESULTS:We propose a new model to detect and classify relations in text, BO-LSTM, that takes advantage of domain-specific ontologies, by representing each entity as the sequence of its ancestors in the ontology. We implemented BO-LSTM as a recurrent neural network with long short-term memory units and using open biomedical ontologies, specifically Chemical Entities of Biological Interest (ChEBI), Human Phenotype, and Gene Ontology. We assessed the performance of BO-LSTM with drug-drug interactions mentioned in a publicly available corpus from an international challenge, composed of 792 drug descriptions and 233 scientific abstracts. By using the domain-specific ontology in addition to word embeddings and WordNet, BO-LSTM improved the F1-score of both the detection and classification of drug-drug interactions, particularly in a document set with a limited number of annotations. We adapted an existing DDI extraction model with our ontology-based method, obtaining a higher F1 score than the original model. Furthermore, we developed and made available a corpus of 228 abstracts annotated with relations between genes and phenotypes, and demonstrated how BO-LSTM can be applied to other types of relations. CONCLUSIONS:Our findings demonstrate that besides the high performance of current deep learning techniques, domain-specific ontologies can still be useful to mitigate the lack of labeled data.
| S-EPMC6323831 | biostudies-literature
Project description:The definition of a precise illumination region is essential in many applications where the electromagnetic field should be confined in some specific volume. By using conventional structures, it is difficult to achieve an adequate confinement distance (or volume) with negligible levels of radiation leakage beyond it. Although metamaterial structures and metasurfaces are well-known to provide high controllability of their electromagnetic properties, this feature has not yet been applied to solve this problem. We present a method of electromagnetic field confinement based on the generation of evanescent waves by means of metamaterial structures. With this method, the confinement volume can be controlled, namely, it is possible to define a large area with an intense field without radiation leakage. A prototype working in the microwave region has been implemented, and very good agreement between the measurements and the theoretical prediction of field distribution has been obtained.
| S-EPMC5123569 | biostudies-literature
Project description:Metamaterials with a Dirac-like cone dispersion at the center of the Brillouin zone behave like an isotropic and impedance-matched zero refractive index material at the Dirac-point frequency. Such metamaterials can be realized in the form of either bulk metamaterials with efficient coupling to free-space light or on-chip metamaterials that are efficiently coupled to integrated photonic circuits. These materials enable the interactions of a spatially uniform electromagnetic mode with matter over a large area in arbitrary shapes. This unique optical property paves the way for many applications, including arbitrarily shaped high-transmission waveguides, nonlinear enhancement, and phase mismatch-free nonlinear signal generation, and collective emission of many emitters. This review summarizes the Dirac-like cone-based zero-index metamaterials' fundamental physics, design, experimental realizations, and potential applications.
| S-EPMC8481486 | biostudies-literature
Project description:This paper presents an algorithm for real-time detection of the heart rate measured on a person's wrist using a wearable device with a photoplethysmographic (PPG) sensor and accelerometer. The proposed algorithm consists of an appropriately trained LSTM network and the Time-Domain Heart Rate (TDHR) algorithm for peak detection in the PPG waveform. The Long Short-Term Memory (LSTM) network uses the signals from the accelerometer to improve the shape of the PPG input signal in a time domain that is distorted by body movements. Multiple variants of the LSTM network have been evaluated, including taking their complexity and computational cost into consideration. Adding the LSTM network caused additional computational effort, but the performance results of the whole algorithm are much better, outperforming the other algorithms from the literature.
| S-EPMC8749621 | biostudies-literature
Project description:Electromagnetic resonators are integrated with advanced elastic material to develop a new type of tunable metamaterial. An electromagnetic-elastic metamaterial able to switch on and off its electromagnetic chiral response is experimentally demonstrated. Such tunability is attained by harnessing the unique buckling properties of auxetic elastic materials (buckliballs) with embedded electromagnetic resonators. In these structures, simple uniaxial compression results in a complex but controlled pattern of deformation, resulting in a shift of its electromagnetic resonance, and in the structure transforming to a chiral state. The concept can be extended to the tuning of three-dimensional materials constructed from the meta-molecules, since all the components twist and deform into the same chiral configuration when compressed.
| S-EPMC4913306 | biostudies-literature
Project description:Metamaterials are artificial structures composed of subwavelength unit cells to control electromagnetic (EM) waves. The spatial coding representation of metamaterial has the ability to describe the material in a digital way. The spatial coding metamaterials are typically constructed by unit cells that have similar shapes with fixed functionality. Here, the concept of frequency coding metamaterial is proposed, which achieves different controls of EM energy radiations with a fixed spatial coding pattern when the frequency changes. In this case, not only different phase responses of the unit cells are considered, but also different phase sensitivities are also required. Due to different frequency sensitivities of unit cells, two units with the same phase response at the initial frequency may have different phase responses at higher frequency. To describe the frequency coding property of unit cell, digitalized frequency sensitivity is proposed, in which the units are encoded with digits "0" and "1" to represent the low and high phase sensitivities, respectively. By this merit, two degrees of freedom, spatial coding and frequency coding, are obtained to control the EM energy radiations by a new class of frequency-spatial coding metamaterials. The above concepts and physical phenomena are confirmed by numerical simulations and experiments.
| S-EPMC5604374 | biostudies-literature
Project description:Quantum bits (qubits) are at the heart of quantum information processing schemes. Currently, solid-state qubits, and in particular the superconducting ones, seem to satisfy the requirements for being the building blocks of viable quantum computers, since they exhibit relatively long coherence times, extremely low dissipation, and scalability. The possibility of achieving quantum coherence in macroscopic circuits comprising Josephson junctions, envisioned by Legett in the 1980's, was demonstrated for the first time in a charge qubit; since then, the exploitation of macroscopic quantum effects in low-capacitance Josephson junction circuits allowed for the realization of several kinds of superconducting qubits. Furthermore, coupling between qubits has been successfully achieved that was followed by the construction of multiple-qubit logic gates and the implementation of several algorithms. Here it is demonstrated that induced qubit lattice coherence as well as two remarkable quantum coherent optical phenomena, i.e., self-induced transparency and Dicke-type superradiance, may occur during light-pulse propagation in quantum metamaterials comprising superconducting charge qubits. The generated qubit lattice pulse forms a compound "quantum breather" that propagates in synchrony with the electromagnetic pulse. The experimental confirmation of such effects in superconducting quantum metamaterials may open a new pathway to potentially powerful quantum computing.
| S-EPMC4941529 | biostudies-literature
Project description:This paper reports a fishnet hyperbolic metamaterial that mimics the electromagnetic properties of magnetically confined plasma. These electromagnetic properties are strongly anisotropic and different from any conventional material, therefore cannot be mimicked by bulk materials. The structure is made of a stack of thin copper grids spaced by Rohacell foam. We numerically and experimentally show that this kind of structuration matches well the properties of a homogeneous plasma. This solution breaks a long-lasting bottleneck and will accelerate the development of high-frequency heating systems to be used in nuclear fusion.
| S-EPMC5895602 | biostudies-literature
Project description:Bulk photopolymerization of a library of synthesized multifunctional azides and alkynes was carried out toward developing structure-property relationships for CuAAC-based polymer networks. Multifunctional azides and alkynes were formulated with a copper catalyst and a photoinitiator, cured, and analyzed for their mechanical properties. Material properties such as the glass transition temperatures (Tg) show a strong dependence on monomer structure with Tg values ranging from 41 to 90 °C for the series of CuAAC monomers synthesized in this study. Compared to the triazoles, analogous thioether-based polymer networks exhibit a 45-49 °C lower Tg whereas analogous monomers composed of ethers in place of carbamates exhibit a 40 °C lower Tg. Here, the formation of the triazole moiety during the polymerization represents a critical component in dictating the material properties of the ultimate polymer network where material properties such as the rubbery modulus, cross-link density, and Tg all exhibit strong dependence on polymerization conversion, monomer composition, and structure postgelation.
| S-EPMC5114021 | biostudies-literature
Project description:This research presents a reverse engineering approach to discover the patterns and evolution behavior of SARS-CoV-2 using AI and big data. Accordingly, we have studied five viral families (Orthomyxoviridae, Retroviridae, Filoviridae, Flaviviridae, and Coronaviridae) that happened in the era of the past one hundred years. To capture the similarities, common characteristics, and evolution behavior for prediction concerning SARS-CoV-2. And how reverse engineering using Artificial intelligence (AI) and big data is efficient and provides wide horizons. The results show that SARS-CoV-2 shares the same highest active amino acids (S, L, and T) with the mentioned viral families. As known, that affects the building function of the proteins. We have also devised a mathematical formula representing how we calculate the evolution difference percentage between each virus concerning its phylogenic tree It shows that SARS-CoV-2 has fast mutation evolution concerning its time of arising. Artificial Intelligence (AI) is used to predict the next evolved instance of SARS-CoV-2 by utilizing the phylogenic tree data as a corpus using Long Short-term Memory (LSTM). This paper has shown the evolved viral instance prediction process on ORF7a protein from SARS-CoV-2 as the first stage to predict the complete mutant virus. Finally, in this research, we have focused on analyzing the virus to its primary factors by reverse engineering using AI and big data to understand the viral similarities, patterns, and evolution behavior to predict future viral mutations of the virus artificially in a systematic and logical way.
| S-EPMC8016547 | biostudies-literature