Search Results (6)

In the task of upper-limb pattern recognition, effective feature extraction, channel selection, and classification methods are crucial for the construction of an efficient surface electromyography (sEMG) signal classification framework. However, existing deep learning models often face limitations due to improper channel selection methods and overly specific designs, leading to high computational complexity and limited scalability. To address this challenge, this study introduces a deep learning network based on channel feature compression—partial channel selection sEMG net (PCS-EMGNet). This network combines channel feature compression (channel selection) and feature extraction (partial block), aiming to reduce the model’s parameter count while maintaining recognition accuracy. PCS-EMGNet extracts high-dimensional feature vectors from sEMG signals through the partial block, decoding spatial and temporal feature information. Subsequently, channel selection compresses and filters these high-dimensional feature vectors, accurately selecting channel features to reduce the model’s parameter count, thereby decreasing computational complexity and enhancing the model’s processing speed. Moreover, the proposed method ensures the stability of classification, further improving the model’s capability of recognizing features in sEMG signal data. Experimental validation was conducted on five benchmark databases, namely the NinaPro DB4, NinaPro DB5, BioPatRec DB1, BioPatRec DB2, and BioPatRec DB3 datasets. Compared to traditional gesture recognition methods, PCS-EMGNet significantly enhanced recognition accuracy and computational efficiency, broadening its application prospects in real-world settings. The experimental results showed that our model achieved the highest average accuracy of 88.34% across these databases, marking a 9.96% increase in average accuracy compared to models with similar parameter counts. Simultaneously, our model’s parameter size was reduced by an average of 80% compared to previous gesture recognition models, demonstrating the effectiveness of channel feature compression in maintaining recognition accuracy while significantly reducing the parameter count. Full article

Enhancing information representation in electromyography (EMG) signals is pivotal for interpreting human movement intentions. Traditional methods often concentrate on specific aspects of EMG signals, such as the time or frequency domains, while overlooking spatial features and hidden human motion information that exist across EMG channels. In response, we introduce an innovative approach that integrates multiple feature domains, including time, frequency, and spatial characteristics. By considering the spatial distribution of surface electromyographic electrodes, our method deciphers human movement intentions from a multidimensional perspective, resulting in significantly enhanced gesture recognition accuracy. Our approach employs a divide-and-conquer strategy to reveal connections between different muscle regions and specific gestures. Initially, we establish a microscopic viewpoint by extracting time-domain and frequency-domain features from individual EMG signal channels. We subsequently introduce a macroscopic perspective and incorporate spatial feature information by constructing an inter-channel electromyographic signal covariance matrix to uncover potential spatial features and human motion information. This dynamic fusion of features from multiple dimensions enables our approach to provide comprehensive insights into movement intentions. Furthermore, we introduce the space-to-space (SPS) framework to extend the myoelectric signal channel space, unleashing potential spatial information within and between channels. To validate our method, we conduct extensive experiments using the Ninapro DB4, Ninapro DB5, BioPatRec DB1, BioPatRec DB2, BioPatRec DB3, and Mendeley Data datasets. We systematically explore different combinations of feature extraction techniques. After combining multi-feature fusion with spatial features, the recognition performance of the ANN classifier on the six datasets improved by 2.53%, 2.15%, 1.15%, 1.77%, 1.24%, and 4.73%, respectively, compared to a single fusion approach in the time and frequency domains. Our results confirm the substantial benefits of our fusion approach, emphasizing the pivotal role of spatial feature information in the feature extraction process. This study provides a new way for surface electromyography-based gesture recognition through the fusion of multi-view features. Full article

Bio-based levulinic acid (LA), furfural (FF), and hydroxymethylfurfural (HMF) represent key chemical intermediates when biorefining biomass resources, i.e., either cellulose, glucose, hexoses, etc. (HMF/LA), or hemicellulose, xylose, and pentose (FF). Despite their importance, their online in situ detection by process analytical technologies (PATs), solubility, and its temperature dependence are seldom available. Herein, we report their solubility and temperature dependence by examining n-hexane, cyclohexane, benzene, toluene, 1,4-dioxane, diethyl ether, dichloromethane, tetrahydrofuran, ethyl acetate, acetone, dimethylformamide, acetonitrile, dimethyl sulfoxide, formic acid, n-butanol, n-propanol, ethanol, methanol, and water. These solvents were selected as they are the most common nonpolar, polar aprotic, and polar protic solvents. Fourier-transform infrared (FTIR) spectroscopy was applied as a fast, accurate, and sensitive method to the examined solutions or mixtures. The latter also enables operando monitoring of the investigated compounds in pressurized reactors. Selected temperatures investigated were chosen, as they are within typical operating ranges. The calculated thermodynamic data are vital for designing biorefinery process intensification, e.g., reaction yield optimization by selective compound extraction. In addition to extracting, upstream or downstream unit operations that can benefit from the results include dissolution, crystallization, and precipitation. Full article

Typical controllers for fed-batch cultivations are based on the estimation and control of the specific growth rate in real time. Biocalorimetry allows one to measure a heat signal proportional to the substrate consumed by cells. The derivative of this heat signal is usually used to evaluate the specific growth rate, introducing noise to the resulting estimate. To avoid this, this study investigated a novel controller based directly on the heat signal. Time trajectories of the heat signal setpoint were modelled for different specific growth rates, and the controller was set to follow this dynamic setpoint. The developed controller successfully followed the setpoint during aerobic cultivations of Saccharomyces cerevisiae, preventing the Crabtree effect by maintaining low glucose concentrations. With this new method, fed-batch cultivations of S. cerevisiae could be reliably controlled at specific growth rates between 0.075 h⁻¹ and 0.20 h⁻¹_, with average root mean square errors of 15 ± 3%. Full article

At specific growth rates above a particular critical value, Crabtree-positive microorganisms exceed their respiratory capacity and enter diauxic growth metabolism. Excess substrate is converted reductively to an overflow metabolite, resulting in decreased biomass yield and productivity. To prevent this scenario, the cells can be cultivated in a fed-batch mode at a growth rate maintained below the critical value, µ_crit. This approach entails two major challenges: accurately estimating the current specific growth rate and controlling it successfully over the course of the fermentation. In this work, the specific growth rate of S. cerevisiae and E. coli was estimated from enhanced on-line biomass concentration measurements obtained with dielectric spectroscopy and turbidity. A feedforward-feedback control scheme was implemented to maintain the specific growth rate at a setpoint below µ_crit, while on-line FTIR measurements provided the early detection of the overflow metabolites. The proposed approach is in line with the principles of Bioprocess Analytical Technology (BioPAT), and provides a means to increase the productivity of Crabtree-positive microorganisms. Full article

In this paper we report the first observation, through X-ray diffraction, of noncovalent uracil–uracil (U–U) dimeric π-stacking interactions in carbon nanotube (CNT)–based supramolecular assemblies. The directionally oriented morphology determined using atomic force microscopy revealed highly organized behavior through π-stacking of U moieties in a U-functionalized CNT derivative (CNT–U). We developed a dispersion system to investigate the bio-inspired interactions between an adenine (A)-terminated poly(3-adeninehexyl thiophene) (PAT) and CNT–U. These hybrid CNT–U/PAT materials interacted through π-stacking and multiple hydrogen bonding between the U moieties of CNT–U and the A moieties of PAT. Most importantly, the U···A multiple hydrogen bonding interactions between CNT–U and PAT enhanced the dispersion of CNT–U in a high-polarity solvent (DMSO). The morphology of these hybrids, determined using transmission electron microscopy, featured grape-like PAT bundles wrapped around the CNT–U surface; this tight connection was responsible for the enhanced dispersion of CNT–U in DMSO. Full article