Small Amount Of Sample Research Articles

Improving the triphenylphosphine/triphenylphosphine oxide (TPP/TPPO)‐based method for the absolute and accurate quantification by FTIR‐ATR of hydroperoxides in oils or lipid extracts

AbstractResearch on natural sources of polyunsaturated fatty acids (PUFAs) for both food and nutraceutical prospects has significantly grown in recent years. Some plant oils and lipid extracts contain carotenoids, xanthophylls, sterols, and/or phenolic compounds that can provoke a lipid hydroperoxides (LOOHs) overestimation when quantifying them using nonselective colorimetric assays. Herein, we have optimized a mid‐infrared method using Fourier‐transform infrared spectroscopy coupled with attenuated total reflectance (ATR) to quantify LOOHs in oils or lipid extracts. This method is based on the conversion of triphenylphosphine (TPP) to triphenylphosphine oxide (TPPO) in the presence of hydroperoxides, with a direct assessment of TPPO levels following the formation of an oil film on the ATR crystal's surface, allowing for a low detection limit of 0.5 mmol of LOOH kg−1. The concentration of oil and TPP, as well as the reaction time, were optimized. It was demonstrated that the presence of pigments, unsaponifiable compounds, phenolics, and/or PUFAs on oils, do not disrupt the analysis. Furthermore, the stoichiometry of the TPP/LOOH reaction was examined, confirming the reliability of the method in detecting various forms of hydroperoxides. An accelerated oxidation study was carried out and the hydroperoxide contents measured using TPP/TPPO were found to be comparable to those obtained using the ferric thiocyanate method. Our method offers a fast, simple, robust, and sensitive approach to accurately quantify hydroperoxides, regardless of the chemical composition of oil or lipid extracts.Practical Application: The hydroperoxide assay method outlined in this study allows for the rapid and straightforward detection of hydroperoxides in pure oil matrices. The method's elevated sensitivity facilitates the early identification of oxidation indicators in oils, especially those with significant carotenoid or xanthophyll contents since these compounds may affect the results when using methods based on colorimetric quantification of hydroperoxides. This accurate, precise, and reproducible approach requires only small quantities of test samples and chemical products, making it well suited to routine application in laboratories and industrial environments.

A method for detecting gene doping in horse sports without DNA extraction

AbstractGene doping is prohibited in horse sports and can involve the administration of exogenous genes, called transgenes, to postnatal animals. Quantitative polymerase chain reaction (qPCR) methods have been developed to detect gene doping; however, these generally require DNA extraction from the plasma prior to qPCR. In this study, we developed two methods, direct droplet digital PCR (ddPCR) and nested ddPCR, to detect the equine erythropoietin (EPO) transgene without DNA extraction. Direct ddPCR used pretreated plasma and PCR to detect the EPO transgene spiked at 10 copies/μL. Nested ddPCR utilised pre‐amplification using nontreated plasma, purification of PCR products and PCR to detect the EPO transgene spiked at 1 copy/μL in plasma. These methods successfully detected the EPO transgene after intramuscular injection into horses. Since each method has different detection sensitivity, the combined use of direct ddPCR for screening and nested ddPCR for confirmation may complement each other and prevent the occurrence of false positives, allowing the reliable detection of gene‐doped substances. One advantage of these methods is the small amount of sample required, approximately 2.2–5.0 μl, owing to the lack of a DNA extraction step. Therefore, these tests could be applied to small volume samples as an alternative to conventional gene doping tests.

Accessing self-diffusion on nanosecond time and nanometre length scales with minute kinetic resolution.

Neutron spectroscopy uniquely and non-destructively accesses diffusive dynamics in soft and biological matter, including for instance proteins in hydrated powders or in solution, and more generally dynamic properties of condensed matter on the molecular level. Given the limited neutron flux resulting in long counting times, it is important to optimize data acquisition for the specific question, in particular for time-resolved (kinetic) studies. The required acquisition time was recently significantly reduced by measurements of discrete energy transfers rather than quasi-continuous neutron scattering spectra on neutron backscattering spectrometers. Besides this reduction in acquisition times, smaller amounts of samples can be measured with better statistics, and most importantly, kinetically changing samples, such as aggregating or crystallizing samples, can be followed. However, given the small number of discrete energy transfers probed in this mode, established analysis frameworks for full spectra can break down. Presented here are new approaches to analyze measurements of diffusive dynamics recorded within fixed windows in energy transfer, and these are compared with the analysis of full spectra. The new approaches are tested by both modeled scattering functions and a comparative analysis of fixed energy window data and full spectra on well understood reference samples. This new approach can be employed successfully for kinetic studies of the dynamics focusing on the short-time apparent center-of-mass diffusion.

Bead-based spontaneous Raman codes for multiplex immunoassay

In the immunoassay process, for fulfilling the need to identify multiple analytes in a small amount of complex sample matrix, it is desirable to develop highly efficient and specific multiplex suspension array technology. Raman coding strategy offers an attractive solution to code the suspension arrays by simply combing narrow spectral bands with stable signal intensities through solid-phase synthesis on the resin beads. Based on this strategy, we report the bead-based spontaneous Raman codes for multiplex immunoassay. The study resulted in superior selectivity of the Raman-encoded beads for binding with single and multiple analytes, respectively. With the use of mixed types of Raman-encoded immunoassay beads, multiple targets in small amounts of samples were identified rapidly and accurately. By confirming the feasibility of bead-based spontaneous Raman codes for multiplex immunoassay, we anticipate this novel technology to be widely applied in the near future.

Boosting active domain adaptation with exploration of samples

Nowadays, the idea of active learning is gradually adopted to assist domain adaptation. However, due to the existence of domain shift, the traditional active learning methods originating from semi-supervised scenarios can not be directly applied to domain adaptation. To solve the problem, active domain adaptation is proposed as a new domain adaptation paradigm, which aims to improve the performance of the model by annotating a small amount of target domain samples. In this regard, we propose an active domain adaptation method named Boosting Active Domain Adaptation with Exploration of Samples (BADA), dividing Active DA into two related issues: sample selection and sample utilization. We design the instability selection criterion based on predictive consistency and the diversity selection criterion. For the remaining unlabeled samples, we design a self-training framework, which screens out reliable samples and unreliable samples through the sample screening mechanism similar to selection criteria. And we adopt respective loss functions for reliable samples and unreliable samples. Experiments show that BADA remarkably outperforms previous active learning methods and Active DA methods on several domain adaptation datasets.

Enhanced Proteomic Coverage in Tissue Microenvironment by Immune Cell Subtype Library-Assisted DIA-MS

Immune cells that infiltrate the tumor microenvironment (TME) play crucial roles in shaping cancer development and influencing clinical outcomes and therapeutic responses. However, obtaining a comprehensive proteomic snapshot of tumor-infiltrating immunity in clinical specimens is often hindered by small sample amounts and a low proportion of immune infiltrating cells in the TME. To enable in-depth and highly sensitive profiling of microscale tissues, we established an immune cell-enriched library-assisted strategy for data-independent acquisition mass spectrometry (DIA-MS). Firstly, six immune cell subtype-specific spectral libraries were established from sorted cluster of differentiation markers, CD8+, CD4+ T lymphocytes, B lymphocytes, natural killer cells, dendritic cells, and macrophages in murine mesenteric lymph nodes (MLNs), covering 7815 protein groups with surface markers and immune cell-enriched proteins. The feasibility of microscale immune proteomic profiling was demonstrated on 1 μg tissue protein from the tumor of murine colorectal cancer (CRC) models using single-shot DIA; the immune cell-enriched library increased coverage to quantify 7419 proteins compared to directDIA analysis (6978 proteins). The enhancement enabled the mapping of 841 immune function-related proteins and exclusive identification of many low-abundance immune proteins, such as CD1D1, and CD244, demonstrating high sensitivity for immune landscape profiling. This approach was used to characterize the MLNs in CRC models, aiming to elucidate the mechanism underlying their involvement in cancer development within the TME. Even with a low percentage of immune cell infiltration (0.25–3%) in the tumor, our results illuminate downregulation in the adaptive immune signaling pathways (such as C-type lectin receptor signaling, and chemokine signaling), T cell receptor signaling, and Th1/Th2/Th17 cell differentiation, suggesting an immunosuppressive status in MLNs of CRC model. The DIA approach using the immune cell-enriched libraries showcased deep coverage and high sensitivity that can facilitate illumination of the immune proteomic landscape for microscale samples.

Crossroads of vibrational (infrared and Raman) spectroscopy and X-ray powder diffraction in identification and characterization of some minerals – advantages and limitations. A review

Many analytical methods have been successfully employed for the study of minerals, in particular, vibrational infrared (FTIR) and Raman spectroscopies and X-ray powder diffraction (XRPD). The advantages of the vibrational spectroscopic techniques for identifying and characterizing minerals include: rapid and versatile use; qualitative and quantitative chemical signatures; distinctive chemical fingerprint of a material; indirect determination of the crystal features (polymorphism, isomorphism, coordination, degree of deformation of structural polyhedra); small sample quantity (area less than 1 μm2 for Raman); wide coverage of 4,000 – 50 cm−1 region in a single scan; in situ and direct measurements without sample preparation; nondestructive use; etc. On the other hand, XRPD is a destructive technique that, depending on the method used and the density of the material, requires from a few micrograms up to around 5 grams of sample quantity for analysis. In spite of that, it is a rapid and powerful technique used in mineral studies with relatively straightforward interpretation of the results. During the last decade, portable X-ray powder diffractometers for the nondestructive analysis of art and archeological materials have been developed along with the portable and hand-held vibrational spectroscopy instrument. Here, some advantages and limitations in the process of the complementary use of FTIR and Raman vibrational spectroscopy and XRPD for identification and characterization of minerals are outlined.

Micro-sample Nd isotopic measurement on reference rock powders via the thermal ionization mass spectrometer Nd+ mode

The Sm–Nd isotopic system has been broadly utilized for dating and tracing in earth and planetary sciences. However, low Nd+ ionization efficiency permanently hinders precise and accurate measurements on small amounts of Nd samples. So far, isotopic measurement of micro-amount Nd sample via thermal ionization mass spectrometry (TIMS) Nd+ mode was mostly performed on Nd reference solutions, but less on natural rock or mineral samples. In the present study, analytical procedures to measure isotopic composition via the TIMS Nd+ mode were designed for natural rock powders by using 1–5 ng pure Nd. The analytical technique was refined through low-blank sample digestion, small-column chemistry for Nd extraction from matrix elements, loading Nd using purified H3PO4 solution, and careful heating of filaments before data acquisition. Measurements on small Nd amounts (1 ng–5 ng) of either the JNdi-1 reference solution or four reference rock powders yielded satisfactory results of 143Nd/144Nd value with respect to precision and accuracy. The analytical results agree well with the recommendation values of the reference materials within analytical errors and the internal precision is better than ±0.005 % (2RSE). When using 1 ng or 2 ng Nd amounts, 144Nd ion flow can remain stable in a signal intensity of >100 mV, and without the assistance of an additional ionization enhancer. Therefore, the analytical procedure, featured by low procedural blank (about 10–22 pg), high Nd+ ionization efficiency, and easy operation, has broad prospects for the application of Nd isotope in single-grain minerals or micro-samples.

A multiplex digital PCR assay for detection and quantitation of porcine circovirus type 2 and type 3.

Porcine circovirus (PCV) has become a major pathogen, causing major economic losses in the global pig industry, and PCV type 2 (PCV2) and 3 (PCV3) are distributed worldwide.We designed specific primer and probe sequences targeting PCV2 Cap and PCV3 Rap anddeveloped a multiplex crystal digital PCR (cdPCR) method after optimizing the primerconcentration, probe concentration, and annealing temperature.The multiplex cdPCR assay permits precise and differential detection of PCV2 and PCV3, with a limit of detection of 1.39 × 101 and 1.27 × 101 copies/reaction, respectively, and no cross-reaction with other porcine viruses was observed. The intra-assay and interassay coefficients of variation (CVs) were less than 8.75%, indicating good repeatability and reproducibility. To evaluate the practical value of this assay, 40 tissue samples and 70 feed samples were tested for both PCV2 and PCV3 by cdPCR and quantitative PCR (qPCR). Using multiplex cdPCR, the rates of PCV2 infection, PCV3 infection, and coinfection were 28.45%, 1.72%, and 12.93%, respectively, and using multiplex qPCR, they were 25.00%, 0.86%, and 4.31%, respectively This highly specific and sensitive multiplex cdPCR thus allows accurate simultaneous detection of PCV2 and PCV3, and it is particularly well suited for applications that require the detection of small amounts of input nucleic acid or samples with intensive processing and complex matrices.

Synthesis of adhesion promoter for lead frame bonding and application of epoxy composite materials

AbstractPolymer adhesion promoters were formulated to improve the adhesion strength of lead frame/epoxy composites. The poly(itaconic acid‐co‐acrylamide) (IAcAAM) was synthesized via free radical polymerization to improve the interfacial adhesion between Ni and Cu lead frames and epoxy composites. IAcAAM was synthesized by setting itaconic acid (IA) and acrylamide (AAM) at molar ratios of 7:3, 5:5, and 3:7. The interfacial adhesion mechanism was investigated by introducing IAcAAM with three different comonomer molar ratios. The adhesion strength between EMC and Cu and Ni substrates to which each IAcAAM was applied was measured using a universal testing machine, and the effect of IAcAAM with various IA:AAM molar ratios on the surface energy of EMC was closely studied. As a result, the adhesion properties of Ni lead frame/epoxy composite and Cu lead frame/epoxy composite were improved as small amounts of all IAcAAM samples were added to the epoxy mixture. We confirmed that IAcAAM with an appropriate molar ratio can be used in the EMC process and has the potential to increase the adhesion of epoxy composites and metal lead frames.Highlights A polymeric adhesive promoter was synthesized to enhance the adhesion strength of lead frame/epoxy composites. Poly(itaconic acid‐co‐acrylamide) (IAcAAM) was copolymerized from itaconic acid and acrylamide for this purpose. The adhesion strength of lead frame/epoxy composites was compared for different IAcAAM monomer contents and with a commercial adhesive promoter. IAcAAM showed superior adhesion strength in the Ni lead frame/epoxy composite material. The highest adhesion strength was achieved at a monomer composition of 5:5.

Rapid analysis of the chemical constituents of traditional Tibetan medicine Sbyor-bzo-ghi-wang using ballpoint electrospray ionization technique.

The chemical constituents of traditional Tibetan medicines (TTM) can be identified using high-performance liquid chromatography and high-resolution mass spectrometry (HPLC-MS/MS) technique. However, the HPLC-MS/MS technique requires the sample to be pretreated and then separated using the specific liquid chromatography method, which is time consuming. This study developed a ballpoint electrospray ionization (BPESI) technique for analyzing the chemical constituents of Sbyor-bzo-ghi-wang. This technique is a simple and inexpensive method for the rapid identification of the chemical constituents of TTMs. After the important parameters of the homemade BPESI device were optimized, the chemical constituents of Sbyor-bzo-ghi-wang were quickly identified without sample pretreatment. The raw data were converted to mzML file using MSConvert and then identified using SIRIUS 5 software. The results showed that 30 compounds were identified from Sbyor-bzo-ghi-wang, namely eight bile acids, six flavonoids, four alkaloids, three amino acids, and nine others. Compared to the ultra-high-performance liquid chromatography-Q/Orbitrap and high-resolution mass spectrometry (UHPLC-Q/Orbitrap HRMS) technique, the BPESI technique identified almost similar types of compounds and also a comparable number of compounds. Compared with the traditional HPLC-MS/MS methods, the BPESI technique does not require complex sample pretreatment and subsequent chromatographic separation steps; also it consumes a small quantity of samples. Therefore, BPESI can be used for the qualitative analysis of the chemical constituents of Sbyor-bzo-ghi-wang.

Synthesis and exploration of the structure and function of artificially synthesized red blood cells

Nowadays, many major diseases and surgical treatments require fresh blood to be injected into patients. However, fresh red blood cells’ storage and typing conditions are very strict, which has led to research on artificial red blood cells. Through conducting research experiments on artificial red blood cells, human blood samples are subjected to a series of steps such as mineralization, filtration, and calcination, ultimately obtaining a small amount of artificial red blood cell samples. This series of research and learning has laid a solid foundation for the demand for large-scale blood production in the medical field, helping to improve the quality and quantity of blood resources in various medical facilities.

Chemical Composition and Antimicrobial Activity of Essential Oils and Hydrosols from Oregano, Sage and Pennyroyal against Oral Pathogens

In this study, the chemical composition of essential oils (EOs) extracted from Origanum vulgare ssp. hirtum Lamiaceae, (oregano), Salvia officinalis Lamiaceae (sage), Mentha pulegium Lamiaceae (pennyroyal), and respective hydrosols (HSs) has been investigated by Gas Chromatography–Mass Spectrometry (GC-MS). The antimicrobial activity was assessed against two oral pathogens: Gram-positive bacterium Streptococcus mutans and the fungus Candida albicans by determination of Minimum Inhibitory Concentration (MIC) and Minimum Bactericidal/Fungicidal concentration (MBC/MFC). Three-fold diluted solutions were dispensed into each well of a 96-well microtiter plate and, after incubation, MIC was determined by visual monitoring. The MBC/MFC was determined by transferring a small quantity of sample contained in each replicate well of the microtiter plates to appropriate culture media using a microplate replicator. The EOs of the tested herbs showed antimicrobial properties, especially the EO oil of O. vulgare, which exerted the highest antimicrobial activity. HSs of S. officinalis and M. pulegium exerted no antimicrobial activity, in contrast to oregano HS, which displayed strong antimicrobial activity. In all cases, a higher number of compounds were detected in EOs than in the corresponding HSs. The major compounds of sage EO were detected to be α-thujone (25.1%), 1,8-cineole (15.8%) and β-pinene (10.0%), while the HS was characterized by the presence of 1,8-cineole (32.6%), borneol (22.6%) and α-thujone (22.4%). Pennyroyal EO and HS consists mainly of pulegone (62.1 and 50.6%, respectively). Carvacrol was the major component present in EO (63%) and HS (97.3%) of oregano, probably contributing to the antimicrobial activity. Further research is needed in order to elucidate the antimicrobial mechanisms of specific compounds present in essential oils and hydrosols of Lamiaceae grown in Greece against oral pathogens.

Microscale Thermophoresis Analysis of Membrane Proteins

Microscale Thermophoresis Analysis of Membrane Proteins

Determination of coumarins in Fructus cnidii by using deep eutectic solvent elution-based carboxymethyl chitosan matrix solid-phase dispersive extraction

Fructus cnidii has been used as a folk medicine for thousands of years. Coumarins are the active compounds behind its therapeutic effect. Thus, the determination of coumarins is necessary for quality evaluation of Fructus cnidii. However, most traditional methods for determining coumarins in natural products are time-consuming, organic solvent-consuming, and are not amenable to the analysis of volatile compounds. Herein, a novel method termed deep eutectic solvent-based carboxymethyl chitosan matrix solid-phase dispersion extraction was proposed and applied to the determination of coumarins in Fructus cnidii, including bergapten, imperatorin, and osthole. For the first time, carboxymethyl chitosan was applied as a dispersant and hexanoic acid-monoethanolamine (a deep eutectic solvent) was used as an eluting solution in the analytical procedure. The deep eutectic solvent dissolved the dispersant (CMC), thereby releasing the analytes and precluding irreversible adsorption. The analytes could be simultaneously concentrated during the elution process. The newly developed method was coupled with HPLC for the determination of coumarins in Fructus cnidii. Several critical parameters affecting the extraction performance, including the molar ratio of the sample to dispersant, grinding time, type of eluting solution, and amount of acid added were investigated. The method was validated by using HPLC. The good linearity (r ≥ 0.9998), a low detection limit, satisfactory precision (intra-day precision is less than 1.88 %, inter-day precision is less than 6.71 %) and accuracy (recovery is 91.3 %-95.0 %) were obtained. Under the optimal conditions, the proposed method had advantages including a simpler extraction process, smaller sample amount (10 mg), less organic reagent (2.4 mL), shorter extraction time (7 min), and no additional concentration process. It is also allowed the analysis of volatile compounds and is environment friendly.

Reliable biological indicator identification and evaluation of tobacco-derived nicotine using an ultra-sensitive gas chromatography-tandem mass spectrometric method.

Several countries have exempted synthetic nicotine from existing regulatory frameworks, resulting in the widespread substitution of synthetic nicotine (SN) in almost all e-cigarette products available. However, it remains uncertain whether the purported synthetic nicotine is indeed genuine SN. There is a need to develop biological indicators and an analytical method that more clearly distinguishes between the two sources. Impurities in neat tobacco-derived nicotine (TDN) were characterized and identified through non-targeted and targeted analysis. Gas chromatography-tandem mass spectrometry (GC-MS/MS) conditions were optimized for detecting biological indicators in e-cigarette products. Nine tobacco-related alkaloids were identified and selected as biological indicators for TDN. A liquid-liquid extraction and GC-MS/MS quantitative method were developed to detect nine biological indicators in e-cigarette products with the limit of quantification ranging from 0.2 to 4.2µg L-1 using 0.5mL of e-liquid. This method was applied to 50 e-cigarette brands purchased in the Korean market. The developed method was able to easily and accurately identify the origin of nicotine even using a small amount of e-liquid sample. It is expected that effective e-cigarette regulation will be possible if the nicotine biological indicator and high-sensitivity analysis method developed in this study are used.

Few-shot detection of surface roughness of workpieces processed by different machining techniques

The traditional deep learning method for detecting workpiece surface roughness relies heavily on a large number of training samples. Also, when detecting the surface roughness of workpieces processed by different machining techniques, it requires a large number of samples of that workpiece to rebuild the model. To address these problems, this paper proposes a few-sample visual detection method for the surface roughness of workpieces processed by different techniques. This method first trains a base model using a relatively large amount of samples from one machining technique, then fine-tunes the model using small amounts of samples from workpieces of different techniques. By introducing contrastive proposal encoding into Faster R-CNN, the model’s ability to learn surface features from small amounts of workpiece samples is enhanced, thus improving the detection accuracy of surface roughness of workpieces processed by different techniques. Experiments show that this method reduces the model’s dependence on training samples and the cost of data preparation. It also demonstrates higher accuracy in surface roughness detection tasks of workpieces processed by different techniques, providing a new approach and insights for few-sample surface roughness detection.

Advanced Sampling Technique in Radiology Free-Text Data for Efficiently Building Text Mining Models by Deep Learning in Vertebral Fracture.

This study aims to establish advanced sampling methods in free-text data for efficiently building semantic text mining models using deep learning, such as identifying vertebral compression fracture (VCF) in radiology reports. We enrolled a total of 27,401 radiology free-text reports of X-ray examinations of the spine. The predictive effects were compared between text mining models built using supervised long short-term memory networks, independently derived by four sampling methods: vector sum minimization, vector sum maximization, stratified, and simple random sampling, using four fixed percentages. The drawn samples were applied to the training set, and the remaining samples were used to validate each group using different sampling methods and ratios. The predictive accuracy was measured using the area under the receiver operating characteristics (AUROC) to identify VCF. At the sampling ratios of 1/10, 1/20, 1/30, and 1/40, the highest AUROC was revealed in the sampling methods of vector sum minimization as confidence intervals of 0.981 (95%CIs: 0.980-0.983)/0.963 (95%CIs: 0.961-0.965)/0.907 (95%CIs: 0.904-0.911)/0.895 (95%CIs: 0.891-0.899), respectively. The lowest AUROC was demonstrated in the vector sum maximization. This study proposes an advanced sampling method, vector sum minimization, in free-text data that can be efficiently applied to build the text mining models by smartly drawing a small amount of critical representative samples.

Supervised Meta-Reinforcement Learning with Trajectory Optimization for Manipulation Tasks

Learning from small amounts of samples with reinforcement learning (RL) is challenging in many tasks, especially in real-world applications, such as robotics. Meta-Reinforcement Learning (meta-RL) has been proposed as an approach to address this problem by generalizing to new tasks through experience from previous similar tasks. However, these approaches generally perform meta-optimization by focusing direct policy search methods on validation samples from adapted policies, thus requiring large amounts of on-policy samples during meta-training. To this end, we propose a novel algorithm called Supervised Meta-Reinforcement Learning with Trajectory Optimization (SMRL-TO) by integrating Model-Agnostic Meta-Learning (MAML) and iLQR-based trajectory optimization. Our approach is designed to provide online supervision for validation samples through iLQR-based trajectory optimization and embed simple imitation learning into the meta-optimization rather than policy gradient steps. This is actually a bi-level optimization that needs to calculate several gradient updates in each meta-iteration, consisting of off-policy reinforcement learning in the inner loop and online imitation learning in the outer loop. SMRL-TO can achieve significant improvements in sample efficiency without human-provided demonstrations, due to the effective supervision from iLQR-based trajectory optimization. In this paper, we describe how to use iLQR-based trajectory optimization to obtain labeled data and then how leverage them to assist the training of meta-learner. Through a series of robotic manipulation tasks, we further show that compared with the previous methods, the proposed approach can substantially improve sample efficiency and achieve better asymptotic performance.

Embedded EEG Feature Selection for Multi-Dimension Emotion Recognition via Local and Global Label Relevance.

Due to the problem of a small amount of EEG samples and relatively high dimensionality of electroencephalogram (EEG) features, feature selection plays an essential role in EEG-based emotion recognition. However, current EEG-based emotion recognition studies utilize a problem transformation approach to transform multi-dimension emotional labels into single-dimension labels, and then implement commonly used single-label feature selection methods to search feature subsets, which ignores the relations between different emotional dimensions. To tackle the problem, we propose an efficient EEG feature selection method for multi-dimension emotion recognition (EFSMDER) via local and global label relevance. First, to capture the local label correlations, EFSMDER implements orthogonal regression to map the original EEG feature space into a low-dimension space. Then, it employs the global label correlations in the original multi-dimension emotional label space to effectively construct the label information in the low-dimension space. With the aid of local and global relevance information, EFSMDER can conduct representational EEG feature subset selection. Three EEG emotional databases with multi-dimension emotional labels were used for performance comparison between EFSMDER and fourteen state-of-the-art methods, and the EFSMDER method achieves the best multi-dimension classification accuracies of 86.43, 84.80, and 97.86 percent on the DREAMER, DEAP, and HDED datasets, respectively.