A GRU-CNN model for auditory attention detection using microstate and recurrence quantification analysis.

Attention as a cognition ability plays a crucial role in perception which helps humans to concentrate on specific objects of the environment while discarding others. In this paper, auditory attention detection (AAD) is investigated using different dynamic features extracted from multichannel electroencephalography (EEG) signals when listeners attend to a target speaker in the presence of a competing talker. To this aim, microstate and recurrence quantification analysis are utilized to extract different types of features that reflect changes in the brain state during cognitive tasks. Then, an optimized feature set is determined by employing the processes of significant feature selection based on classification performance. The classifier model is developed by hybrid sequential learning that employs Gated Recurrent Units (GRU) and Convolutional Neural Network (CNN) into a unified framework for accurate attention detection. The proposed AAD method shows that the selected feature set achieves the most discriminative features for the classification process. Also, it yields the best performance as compared with state-of-the-art AAD approaches from the literature in terms of various measures. The current study is the first to validate the use of microstate and recurrence quantification parameters to differentiate auditory attention using reinforcement learning without access to stimuli.

Thermogravimetric and thermo-kinetic analysis of sugarcane bagasse pith: a comparative evaluation with other sugarcane residues.

In this study, thermogravimetric and thermo-kinetic analysis of sugarcane bagasse pith (S.B.P.) were performed using a robust suite of experiments and kinetic analyses, along with a comparative evaluation on the thermo-kinetic characteristics of two other major sugarcane residues, namely sugarcane straw (S.C.S.) and sugarcane bagasse (S.C.B.). The thermogravimetric analysis evaluated the pyrolysis behavior of these residues at different heating rates in a nitrogen atmosphere. The Kissinger, advanced non-linear isoconversional (ANIC), and Friedman methods were employed to obtain effective activation energies. Moreover, the compensation effect theory (CE) and combined kinetic analysis (CKA) were used to determine the pre-exponential factor and pyrolysis kinetic model. Friedman's method findings indicated that the average activation energies of S.C.S., S.C.B., and S.B.P. are 188, 170, and 151 kJ/mol, respectively. The results of the ANIC method under the integral step Δα = 0.01 were closely aligned with those of the Friedman method. The CKA and CE techniques estimated ln(f(α)Aα) with an average relative error below 0.7%. The pre-exponential factors of S.C.S., S.C.B., and S.B.P. were in the order of 1014, 1012, and 1011 (s-1), respectively. From a thermodynamic viewpoint, positive ∆G* and ∆H* results provide evidence for the non-spontaneous and endothermic nature of the pyrolysis process, indicating the occurrence of endergonic reactions.

Energy consumption prediction in an office building by examining occupancy rates and weather parameters using the moving average method and artificial neural network.

Occupancy rate refers to the level of usage and presence of individuals within a building or a specific space. This factor can have a significant impact on building energy consumption. When the occupancy rate in a building is high, naturally, energy consumption also increases. This correlation might be due to the increased use of lighting, heating, and cooling, higher numbers of electrical and electronic devices, and similar factors associated with the presence of people in the building. One of the modern methods in the energy field involves empirically utilizing occupancy monitoring tools in buildings and analyzing the relationship between such utilization and building energy consumption through artificial neural network tools. In this research, a camera sensitive to entry and exit was installed at the entrance of an office building in Tehran, Iran. By doing so, the rate of entry and exit was accurately monitored. In the next stage, by investigating the impact of this entry and exit rate on the building's energy consumption, the energy consumption amount was predicted using an artificial neural network and a statistical method (moving average). The results indicate errors of 9.8 and 4.5 for the respective methods, highlighting that the artificial neural network yields the most accurate outcomes. Moreover, the study's findings suggest a direct correlation: as occupancy rates increase, the predicted energy consumption values also rise.

Investigation of chalcopyrite removal from low-grade molybdenite using response surface methodology and its effect on molybdenum trioxide morphology by roasting.

In this research, purification of molybdenite concentrate (MoS2) using a nitric acid leaching process was employed for the improvement of molybdenum trioxide morphology during oxidative roasting in an air atmosphere. These experiments were performed using 19 trials designed with response surface methodology and three effective parameters being temperature, time, and acid molarity. It was found that the leaching process reduced the chalcopyrite content in the concentrate by more than 95%. The influence of chalcopyrite elimination and roasting temperature on the morphology and fiber growth of the MoO3 was also investigated by SEM images. Copper plays an important role in controlling the morphology of MoO3 and its decrease led to enhancing the length of quasi-rectangular microfibers from less than 30 µm for impure MoO3 up to several centimeters for purified MoO3.

MLP-RL-CRD: diagnosis of cardiovascular risk in athletes using a reinforcement learning-based multilayer perceptron.

Objective.Pre-participation medical screening of athletes is necessary to pinpoint individuals susceptible to cardiovascular events.Approach.The article presents a reinforcement learning (RL)-based multilayer perceptron, termed MLP-RL-CRD, designed to detect cardiovascular risk among athletes. The model underwent training using a publicized dataset that included the anthropological measurements (such as height and weight) and biomedical metrics (covering blood pressure and pulse rate) of 26 002 athletes. To address the data imbalance, a novel RL-based technique was adopted. The problem was framed as a series of sequential decisions in which an agent classified a received instance and received a reward at each level. To resolve the insensitivity to the initialization of conventional gradient-based learning methods, a mutual learning-based artificial bee colony (ML-ABC) was proposed.Main Results.The model outcomes were validated against positive (P) and negative (N) ECG findings that had been labeled by experts to signify individuals 'at risk' and 'not at risk,' respectively. The MLP-RL-CRD approach achieves superior outcomes (F-measure 87.4%; geometric mean 89.6%) compared with other deep models and traditional machine learning techniques. Optimal values for crucial parameters, including the reward function, were identified for the model based on experiments on the study dataset. Ablation studies, which omitted elements of the suggested model, affirmed the autonomous, positive, stepwise influence of these components on performing the model.Significance.This study introduces a novel, effective method for early cardiovascular risk detection in athletes, merging reinforcement learning and multilayer perceptrons, advancing medical screening and predictive healthcare. The results could have far-reaching implications for athlete health management and the broader field of predictive healthcare analytics.

A systematic study on hydrogen bond interactions in sulfabenzamide: DFT calculations of the N-14, O-17, and H-2 NQR parameters.

A systematic computational study was carried out to characterize the hydrogen bond, HB, interactions of sulfabenzamide crystal structure by DFT calculations of electric field gradient, EFG, tensors at the sites of 14N, 17O, and 2H nuclei. The computations were performed with the B3LYP and B3PW91 DFT methods and 6-311+G and 6-311++G* standard basis sets using the Gaussian 98 package. To perform the calculations, a hydrogen-bonded heptameric cluster of sulfabenzamide was created by X-ray coordinates where the hydrogen atom positions were optimized and the EFG tensors were calculated for the target molecule. Additional optimization and EFG calculations were also performed for crystalline monomer and an isolated gas-phase sulfabenzamide. The calculated EFG tensors were converted to the experimentally measurable nuclear quadrupole resonance, NQR, parameters: quadrupole coupling constant, C(Q), and asymmetry parameter, eta(Q). The results reveal that the geometrical and NQR parameters of the optimized isolated gas-phase and crystalline phase are different. In addition, the difference between the calculated NQR parameters of the monomer and the target molecule shows how much H-bonding interactions affect the EFG tensors of each nucleus. The evaluated NQR parameters reveal that due to the contribution of the target molecule to N-H...O and C-H...O hydrogen bond interactions, the EFG tensors at the sites of N1, O3 and H1 undergo significant changes from monomer to the target molecule in cluster. These features reveal the major role of N-H...O type intermolecular HBs in cluster model of sulfabenzamide which the presence of these interactions can lead to polymorphism directly related to the drug activity and related properties.