Hardness Parameters Research Articles

Enhancing the quality and antioxidant properties of beef burgers with dried Hibiscus sabdariffa L. leaves: Characterization and sensory evaluation

The demand for natural antioxidants has increased as research efforts aim to identify various plant sources that could be considered sustainable alternatives to preserve the quality of meat and its derivatives. This work intended to develop beef burger formulations with different percentages of dried Hibiscus sabdariffa L. leaves and characterize them through physical, chemical, and sensory analysis. Four beef burger formulations were prepared: F1 - with no added antioxidant, F2 - with added commercial antioxidant, F3 - with the addition of 4 g/kg of hibiscus leaves, and F4 - with the addition of 8 g/kg of hibiscus leaves, which were subjected to analysis of proximate composition, bioaccessibility of phenolic compounds, total antioxidant activity, morphology, texture profile, and sensory profile, as perceived by students from a public school. The addition of hibiscus caused an increase in the ash (∼5%) and polyphenol content (∼15%) of the samples without the addition of antioxidants, reduced the redness of the samples but did not affect the food matrix observed in the micrographs or the hardness, gumminess, and chewiness parameters. The in vitro digestion process improved the accessibility of polyphenols (211.96%–248.31%) and the total antioxidant activity by the ABTS•+ radical (224.77%–281.63%) in all samples. From a sensory standpoint, the samples showed global acceptance levels ranging from “indifferent” to “loved it”. Adding hibiscus was linked to the terms “greenish brown color” and “aftertaste”. The F3 showed promising results in physical and chemical analyses and less interference in its sensorial attributes from adding hibiscus leaves.

Synbiotic-Glyconutrient Additive Reveals a Conducive Effect on Growth Performance, Fatty Acid Profile, Sensory Characteristics, and Texture Profile Analysis in Finishing Pig.

This study aims to investigate the effect of synbiotic-glyconutrients (SB-GLN) additive on growth performance, fatty acid profile, sensory characteristics, and texture profile analysis in finishing pig. Landrace × Yorkshire ♀ × (Duroc ♂) (n = 60) pigs with average body weight of 54.88 ± 1 kg were allocated into one of three dietary treatment groups in a complete randomized block design with four replicates of five pigs (two barrows and three gilts) per pen. The test treatments (TRT) were CON-corn-soybean meal basal diet; TRT 1-CON+ 0.25% SB-GLN; and TRT 2-CON + 0.5% SB-GLN. SB-GLN contains 1 × 107 CFU/g each of: L. plantarum, B. subtilis, and S. Cerevisiae, and 5% yeast cell wall β-Glucans (from S. Cerevisiae), and 14% of glyconutrients (N-acetylglucosamine, D-xylose, and Fucose). Pigs fed SB-GLN supplement showed linearly increased (p < 0.05) body weight, daily gain, and daily feed at the end of week 5, 10, and the overall experimental period. In addition, G:F showed a tendency to decrease (p < 0.1) at the end of week 10 and the overall experimental period. In addition, pigs that received a graded level of SB-GLN showed a tendency to increase (p < 0.1) their longiness muscle area and decreased (p < 0.05) cooking loss. The sensory results of pork belly (tenderness and juiciness) and loin (flavor) meat, and the texture profile analysis parameters of hardness 1, cohesiveness, and gumminess (belly), and hardness 2, chewiness, and springiness (loin) meat were linearly higher (p < 0.05) in the SB-GLN group. The values of fatty acid like butyric acid, caproic acid, undecylic acid, tridecylic acid, myristic acid, pentadecyclic acid, palmitic acid, margaric acid, stearic acid, eicosapentaenoic acid, and lignoceric acid were higher in pork belly fat of the SB-GLN-treated group compared to CON. Moreover, pigs that received SB-GLN exhibited higher crude fat and lauric acid, myristic acid, pentacyclic acid, palmitic acid, margaric acid, Octadecanoic acid, Oleic acid, linoleic acid, and eicosapentaenoic acid FA profiles in belly-lean meat. Also, the FA profile of the SB-GLN-treated group loin-lean meat showed increased lauric acid, myristic acid, palmitic acid, margaric acid, stearic acid, oleic acid, linoleic acid, alpha-linoleic acid, and eicosapentaenoic acid. The SB-GLN-treated group pork belly fat, belly lean meat, and loin-lean meat showed linearly increased docosahexaenoic acid, nervonic acid, omega 3, omega 6, ω-6: ω-3, Σ saturated FA, Σ un-SFA, Σ mono-USFA, Σ poly-USFA, MUFA/SFA, and PUFA/SFA. Therefore, we infer that the inclusion of 0.5% SB-GLN additive to finishing pig diet would be more beneficial to enhance their performance, and to increase the essential FA profile of pork meat for human consumption.

PROCESSING OPTIMIZATION OF CRAFT CHEESE CAMEMBERT TYPE

The article substantiates the technology of craft cheese Camembert type with the use of goat milk and dried plant products for restaurants implementation. Differences in the biological value and technological properties of goat and cow milk were analyzed. It is shown that goat milk contains more protein, fat, lactose, potassium, calcium, magnesium and iodine in comparison to cow milk. The density of goat milk is about 1.033 g∕cm3, while for cow milk this indicator varies within 1.024 and 1.037 g∕cm3. Goat milk compared to cow milk contains more retinol, thiamine, ascorbic acid and less such an allergens as αs1-casein and α-lactoalbumin, that proofs an increased biological value of goat milk. Using the methods of experimental and statistical planning, a quadratic model was obtained that adequately describes the dependence of the cheese hardness parameter on the the milk fermentation temperature and the dosage of goat milk in the recipe of Camembert type cheese. As a result of the minimization of the cheese hardness parameter according to the obtained model, it was determined that the of goat milk should be added in the recipe of craft cheese Camembert type in an amount of 35.6% of cow milk content, and the fermentation of milk should be carried out at a temperature of 36°C. These parameters allow to obtain cheese with the hardness of 13.2%. A technological scheme for the production of craft cheese Camembert type was developed, that attaches mixing of goat milk and cow milk, fermentation of milk at a temperature of 36°C, as well as the addition of dried plant products, such as thyme, rosemary and tomatoes at the stage of a clot processing . Based on the results of the evaluation of the organoleptic quality indicators of the cheese, it was found that it has the correct shape, rich and balanced taste, as well as a soft and delicate consistency, which allows to recommend it for implementation into restaurants production.

Investigation of Geometric and Hardness Parameters of Tank Track Grooves Equipped on Photovoltaic Cleaning Robot

The utilization of photovoltaic (PV) cleaning robots has proven to be an effective method for maintaining the conversion efficiency of utility-scale PV power plants by mitigating the impact of dust accumulation. However, ensuring the safe operation of these robots, resembling tanks in appearance, particularly in wet working conditions, relies heavily on their adherence to PV panels. This study focuses on assessing the slip resistance of candidate materials coated on endless polyurethane timing belts, which are equipped on PV cleaning robots to enable the efficient cleaning of uneven and misaligned PV arrays. A novel apparatus is proposed to evaluate the coefficient of static friction (COSF) of slip specimens, considering factors such as outsole patterns, area density, and shore hardness. The results highlight the significant influence of shore hardness and area density on the slip resistance of the specimens. Based on the findings, it is recommended to design track grooves with hexagon or zigzag patterns and maintain a low area density (e.g., 0.44 g·mm−2) to ensure the safe operation of PV cleaning robots, irrespective of the working conditions they encounter. By addressing the slip resistance challenge, this research contributes to the overall efficiency and reliability of PV cleaning robots, enhancing their performance in maintaining clean and optimal PV panel surfaces.

The Physicochemical Properties and Melting Behavior of Ice Cream Fortified with Multimineral Preparation from Red Algae.

Ice cream is a popular frozen food consumed worldwide throughout the year. However, as a thermally unstable product, it requires proper cold chain management. Thermal fluctuations alter the physicochemical properties of ice cream and reduce its quality. This study was conducted to evaluate the physicochemical and sensory properties of ice cream containing different amounts (0.5; 0.8; 1.0%) of a multimineral preparation from Atlantic red algae. The effect of thermal shock on the quality of ice cream after preparation and 90 days of frozen storage was studied. The addition of a multimineral component slightly increased the freezing and glass transition temperatures of the ice cream. The overrun of the ice cream ranged from 48.55 to 52.78% and decreased with the frozen storage time, but the samples with 0.8 and 1.0% mineral content had the most stable overrun in terms of storage time and thermal fluctuations. Ice cream stored for both 7 and 90 days showed a similar melting behavior, although a shift in the melting curves was observed after long frozen storage. The samples exposed to the thermal treatment had lower melting rates by 39.2-59.9% and 55.2-65.4% for 7-day and 90-day stored ice cream, respectively. The hardness parameters of the ice cream did not change significantly under the conditions applied, so the fragility of the ice cream and its fluffiness did not seem to be affected. The organoleptic evaluation showed that ice cream with a mineral content of 0.8% was the most acceptable in terms of taste, texture, and overall acceptability. The applied mineral and sucrose content ratios did not alter the main physicochemical and organoleptic parameters, but significantly affected the nutrient density of the ice cream.

Clinical and Cephalometric Changes in Long-face Patients With Retrogenia Following the Novel B-genioplasty Technique.

The chin plays a fundamental role in facial attractiveness. Accordingly, several surgical procedures have been proposed for chin contouring. This study aimed to introduce "B-genioplasty" as a novel genioplasty technique and assess the clinical and cephalometric changes in long-face patients with retrognathia following the novel B-genioplasty. This is a cross-sectional study. Subjects with class I occlusion increased lower facial height and retrognathia underwent B-genioplasty in this prospective experimental study. Soft and hard tissue cephalometric measurements were made, and the depth of the notch in the lower border of the mandible was measured on lateral cephalograms at 6 months postoperatively and compared with baseline. Patient satisfaction with the outcome was also assessed. Seventeen subjects were studied. Ten (58%) patients underwent rhinoplasty plus genioplasty. All patients were satisfied with the outcome. The step/notch in the lower border of the mandible was not detected in clinical examination in any patient. The mean notch depth was 0.5±0.83 mm on lateral cephalograms. The change in all soft and hard tissue cephalometric parameters was significant at 6 months postoperatively compared with baseline ( P <0.001 for all). B-genioplasty caused satisfactory clinical and cephalometric changes in long-face patients with retrognathia, including prevention of step/notch in the inferior border of the mandible, significant reduction of gonial angle, correction of the mandibular plane, and reduction of lower facial height.

Investigation of the Dynamics of Processing Parameters in Hard Turning Modes

One of the types of finishing operations of machining the surface of products made of metal materials is hard turning, which is carried out at elevated turning modes. When processing steels hardened to high hardness, in processing modes with the ma-nifestation of all the features characterizing hard turning, a lot of heat is released, which turns into chips during turning, while the values of the cutting force decrease, which is associated with the softening of the active surface layers of the processed material and its subsequent plasticization, which leads to a decrease in cutting forces in the contact zone. A significant interest both from a scientific and practical point of view is the issue of changing the cutting temperature and processing parameters in hard turning modes. The paper presents experimental research results and empirical dependences of the radial component of the cutting force and temperature on cutting speed and feed per revolution, at a fixed cutting depth in the modes of hard turning of hardened steel. The research has been carried out on an experimental installation specially designed for this purpose, consisting of a lathe and a stand for recording, monitoring and analyzing dynamic processes (in the form of a signal from the cutter) during turning. A cutter with replaceable ceramic cutting plates was used as a test tool, and the material of the workpieces for the research was HVG steel, subjected to heat treatment for a hardness category of 55 HRC. Computer and mathematical processing of the results obtained using the least squares method was carried out, which allowed us to derive calculation formulas for determining the values of the studied factors. The optimal values of the factors corresponding to the minimum value of the radial component of the force and the cutting temperature are determined.

Determination of Groundwater Quality Index in Rural Area: The Case of Bartın City

The aim of the study is to determine the change of groundwater quality in rural areas in rainy and dry periods with respect to physicochemical parameters. pH, total dissolved solids, electrical conductivity, nitrate, sulfate, phosphate hardness, chloride, turbidity and color parameters were investigated. The water quality index (WQI) is widely used for detecting and evaluating water pollution. Water quality index was determined to be 35 and 32 in rainy and dry periods for drinking water. It was also calculated to be 37 for the rainy and dry periods according to the irrigation water limit values. As a result, since WQI

Characteristics of Soft Bone Milkfish (Chanos chanos) with Different Boiling Times

Soft bone milkfish is a modification technique by softening the fish bone using high temperature. The advantage of this method is lower the risk of bone chocking while consuming the milkfish. The purpose of this study was to determine the effect of boiling time on the characteristics of soft durian milkfish and the best boiling time favored by the community. The processing method begins with weeding the raw materials and washing them clean, arranging them in a bamboo basket covered with floating bamboo leaves. Boiling is done in a large cauldron with wood fuel and filled with water one third of the height of the casserole with ground spices. Conventional cooking is done for 8 hours. The research design used was a pressure cooker for 2 hours (control), conventional cooking for 6 hours, 7 hours and 8 hours. Cooking time did not affect protein content and moisture content, but did affect the levels of amino acid profiles, bone hardness and hedonic parameters of appearance, smell, taste and texture. The results showed that the best treatment was conventional cooking for 7 hours which had an appearance value of 8.57 ± 0.50, odor 8.67 ± 0.48, taste 8.47 ± 0.51, texture 8.53 ± 0.51, protein content 31.12 ± 0.84%, moisture content 64.48 ± 2.39%, essential amino acid profile content 135.521 mg/1000g, profile content non-essential amino acids 132.785 mg/1000g, bone hardness 44.12 ± 14.11N.

Adaptive Hard Parameter Sharing Method Based on Multi-Task Deep Learning

Multi-task learning (MTL) improves the performance achieved on each task by exploiting the relevant information between tasks. At present, most of the mainstream deep MTL models are based on hard parameter sharing mechanisms, which can reduce the risk of model overfitting. However, negative knowledge transfer may occur, which hinders the performance improvement achieved for each task. In this paper, for situations when multiple tasks are jointly trained, we propose the adaptive hard parameter sharing method. On the basis of the adaptive hard parameter sharing method, the number of nodes in the network is dynamically updated by setting a continuous gradient difference-based sign threshold and a warm-up training iteration threshold through the relationships between the parameters and the loss function. After each task fully utilizes the shared information, adaptive nodes are used to further optimize each task, reducing the impact of negative migration. By using simulation studies and instance analyses, we demonstrate theoretical proof that the performance of the proposed method is better than that of the competing method.

Effects of storage and toothbrush simulation on Martens hardness of CAD/CAM, hand-cast, thermoforming, and 3D-printed splint materials

ObjectivesTo investigate Martens hardness parameters of splint materials after storage in liquids and toothbrush simulation.Materials and methodsTen specimens per material and group were fabricated (hand-cast CAST, thermoformed TF, CAD/CAM-milled CAM, 3D-printed PS, PL, PK, PV), stored in air, water, coffee, red wine, and cleaning tablets and investigated after fabrication, 24 h, 2- and 4-week storage or toothbrushing. Martens hardness (HM), indentation hardness (HIT), indentation modulus (EIT), the elastic part of indentation work (ηIT), and indentation creep (CIT) were calculated (ISO 14577-1). Statistics: ANOVA, Bonferroni post hoc test, between-subjects effects, Pearson correlation (α = 0.05).ResultsHM varied between 30.8 N/mm2 for PS (water 4 weeks) and 164.0 N/mm2 for CAM (toothbrush). HIT values between 34.9 N/mm2 for PS (water 4 weeks) and 238.9 N/mm2 for CAM (toothbrush) were found. EIT varied between 4.3 kN/mm2 for CAM (toothbrush) and 1.8 kN/mm2 for PK (water 2 weeks). ηIT was found to vary between 16.9% for PS (water 4 weeks) and 42.8% for PL (toothbrush). CIT varied between 2.5% for PL (toothbrush) and 11.4% for PS (water 4 weeks). The highest impact was identified for the material (p ≤ 0.001).ConclusionsStorage and toothbrushing influenced Martens parameters. The properties of splints can be influenced by the choice of materials, based on different elastic and viscoelastic parameters. High HM and EIT and low CIT might be beneficial for splint applications.Clinical relevanceMartens parameters HM, EIT, and CIT might help to evaluate clinically relevant splint properties such as hardness, elasticity, and creep.

Properties of Yogurts Enriched with Crude Polysaccharides Extracted from Pleurotus ostreatus Cultivated Mushroom

Increasingly, consumers are looking for products with specific nutritional and health-promoting properties. The answer of the producers for this demand is fortified food. The raw material that can be used to enrich food is, among others, mushrooms. Crude water soluble polysaccharides (cWSP) were isolated from fruiting bodies of Pleurotus ostreatus (oyster) mushroom. Chemical analysis showed that they consisted mainly of carbohydrates (~61%), protein (~9%) and phenolics (~0.8%). The isolated cWSP were used to obtain enriched cow milk set yogurts. cWSP were added at the concentration of 0.1%, 0.2%, 0.3%, 0.4% and 0.5%, and milk containing no cWSP was prepared as the control. All of the variants were fermented via applying two commercially available culture starters. The addition of cWSP led to a drop in pH in the case of one starter culture. Also, the decline in total soluble solids (TSS) content was higher where cWSP was used for the enrichment. Texture profile analysis (TPA) revealed that parameters of hardness and gumminess increased along with the concentration of cWSP (reaching values approximately 7–8 times higher, compared to the control). A significant increase in syneresis level (proportional to cWSP concentration and ranging from ~10% to ~50%) was also observed after the fermentation. Fortifying milk with cWSP led to a slight increase in antioxidant capacity in FRAP assay (up to ~12%) and ABTS assay (up to ~23%). The results demonstrate that using cWSP to enrich set-type yogurts is fairly limited.

Optimization of process parameters effects on mechanical properties of selective laser melted iron samples

AbstractSelective laser melting is an additive manufacturing technology whereby parts are built through selective melt of a powder bed by a laser beam. This technology is used to produce net shape parts in 3D printers. In this study, the effect of selective laser melting parameters on the mechanical properties (ultimate tensile strength and hardness) of iron has been experimentally extracted, and the optimum levels of the parameters have been determined for optimum mechanical properties using signal‐to‐noise analysis. The selected parameters include the laser power, the laser scanning speed, and the laser hatch distance, and the design of experiments was done by the Taguchi method. The main effects of factors and interactions were considered in this paper. The results show that the optimum parameter levels for the ultimate tensile strength include the laser power of 200 W, the laser scanning speed of 600 mm/s, and the hatch distance of 70 μm. The optimal parameters for hardness are similar to the maximum tensile strength, except that the optimum hatch distance is 60 μm.

A GIS Based Study to Investigate of the Ecological Impacts of Successive Dams on Surface Water Quality: A Review of the Sakarya Basin, Türkiye

The present study aimed to assess the status of successive dams on surface water quality, especially in reducing organic pollution levels, and their effects on the ecological life were investigated. Water samples were collected seasonally from three dam’s monitoring stations. Temperature, pH, dissolved oxygen, electrical conductivity, turbidity, ammonium nitrogen, nitrite nitrogen, nitrate-nitrogen, sulfate, total phosphorus, chemical oxygen demand, biochemical oxygen demand and total hardness parameters were measured in surface waters. The Geographic Information System (GIS) was used to produce thematic maps of the investigated parameters. Stations used as point feature layers were mapped using quantitative labeling via symbology. The results were compared with Turkish Regulation, 2015 and EC Directive on the quality of fresh waters needing protection or improvement to support fish life (2006) guidelines. According to the results of the analysis, Dams that are located one after the other contribute to the partial cleaning of the water by resting. Sarıyar, Gökçekaya, and Yenice Dams were built on the Sakarya River, which is one of Türkiye’s biggest and most significant river systems, for electric supply and utility of water for irrigation and fishery activities. Gökçekaya Dam Lake was built between Sarıyar Dam and Yenice Dam, which are situated in the same line of Sakarya River. Gökçekaya and Yenice Dams are fed by Sarıyar Dam Lake and do not receive fresh water. The fact that the three dams are established one after the other aids in the water’s resting and the aids in removing the suspended solids/suspended load. But, this research show that domestic, industrial, agricultural wastes carried by the branches of the rivers caused algal blooms in Dam lakes, especially which are not fed with fresh water. Due to organic pollution the decrease in dissolved oxygen, especially from time to time, negatively affects the ecological life in the lakes.

Ultra-thin 2D Fe3GaTe2 rare-earth free permanent magnet at finite temperatures

Despite extensive studies on permanent magnets (PMs) in bulk-type materials, the possibility of PMs in two-dimensional (2D) materials is barely explored so far. In this work, we systematically investigate temperature dependent magnetic properties of 2D trilayer and four-layer Fe3GaTe2 systems based on the first principle calculations. The calculated Curie temperature (T C) in both trilayer and four-layer structures are 340 K–352 K. Both systems have perpendicular magnetic anisotropy, and the uniaxial anisotropy constant is monotonically decreased with increasing temperature. At 300 K, the 2D Fe3GaTe2 has a coercive field of 0.34 T in the trilayer and it becomes 0.44 T in the four-layer. Besides, both systems have a magnetic hardness parameter κ larger than 1 even at 300 K. We also obtain a maximum energy product (BH)max of 24 kJ m−3 in the trilayer, and it is further increased to 26 kJ m−3 in the four-layer at 300 K. Nonetheless, these (BH)max are decreased by more than two times with including the demagnetization factor. Overall, we obtain that 2D Fe3GaTe2 at 2–3 nm (trilayer and four-layer) thickness possesses the same scale of coercive field and maximum energy product of well-known bulk ferrite PM. Our findings may indicate that the atomically thin 2D system can be a potential rare-earth-free PM for small-scale device applications.

Coastal aquifer water quality for domestic supply of small communities in the semiarid Potiguar - Brazilian Northeast

This study analyzed groundwater quality used to supply small communities in the coastal zone of the Brazilian Northeast, identifying the risks from domestic use of water, and potential contaminants from domestic effluents. We realized analyses of 24 chemical and physical parameters in 15 samples of water from shallow wells. The data was processed through quality modeling, descriptive statistics, principal component analysis, and geoespacialization through kriging. The groundwater used by communities showed some limitations to domestic use. Parameters of Br, hardness, Na+, NO3−, Cl−, SO4−, and turbidity exceed the allowed for Brazilian Health Minister (Ministerial Order n. 888/2021). Furthermore, TDS values present low water quality, predominantly brackish and saltwater, most salinity waters are related to depth and sea proximity. Depth of saturated zone probably is a determinant in nitrate concentrations, comparing the study areas. Water self-supply is considered a risk by health because inferior natural quality of water used and contaminations from domestic sewage determine concentrations of some parameters above of legislation.

Comparative efficacy of different tenderizing agents and time on Physicochemical textural and organoleptic properties of squid (Todarodes pacificus) muscle.

This study aimed to investigate the chemical, physical, textural, and sensory properties of the squid (Todarodes pacificus) muscle during different tenderization agents (yeast, milk, and mineral water) and times (3, 6, 12, and 24 h). The results of the analyses showed that different treatments and their durations affected the dry matter, ash, lipid, and crude protein content of the squid. According to sodium dodecyl sulfate-polyacrylamide gel electrophoresis results, it was observed that there was a slight decrease in band intensities based on different treatments and their durations. However, no significant changes were observed in myosin actin and paramyosin bands. It was found that the hardness (359.7 N), cohesiveness (0.63), and gumminess (233.2 N) parameters increased at the 6th hour, decreased at the 12th hour, and increased again at the 24th hour of the treatment. According to the scanning electron microscopy results, the most affected groups by the application and duration were found to be the mineral water group at the 12th and 24th hours, and the milk group at the 24th hour. Fibers in all marinated squid muscles were observed to spring significantly more compared to the positive and negative control groups. The taste score was found to be significantly higher in the group treated with yeast for 3 h and it was evaluated by the panelists as the most delicious squid among all of the groups. As a result of the study, it was determined that the chemical, physical, and sensory properties of squid could be improved by marinating with yeast, milk, and mineral water for different durations.

Cross-dataset transfer learning for motor imagery signal classification via multi-task learning and pre-training

Objective. Deep learning (DL) models have been proven to be effective in decoding motor imagery (MI) signals in Electroencephalogram (EEG) data. However, DL models’ success relies heavily on large amounts of training data, whereas EEG data collection is laborious and time-consuming. Recently, cross-dataset transfer learning has emerged as a promising approach to meet the data requirements of DL models. Nevertheless, transferring knowledge across datasets involving different MI tasks remains a significant challenge in cross-dataset transfer learning, limiting the full utilization of valuable data resources. Approach: This study proposes a pre-training-based cross-dataset transfer learning method inspired by Hard Parameter Sharing in multi-task learning. Different datasets with distinct MI paradigms are considered as different tasks, classified with shared feature extraction layers and individual task-specific layers to allow cross-dataset classification with one unified model. Then, Pre-training and fine-tuning are employed to transfer knowledge across datasets. We also designed four fine-tuning schemes and conducted extensive experiments on them. Main results: The results showed that compared to models without pre-training, models with pre-training achieved a maximum increase in accuracy of 7.76%. Moreover, when limited training data were available, the pre-training method significantly improved DL model’s accuracy by 27.34% at most. The experiments also revealed that pre-trained models exhibit faster convergence and remarkable robustness. The training time per subject could be reduced by up to 102.83 s, and the variance of classification accuracy decreased by 75.22% at best. Significance: This study represents the first comprehensive investigation of the cross-dataset transfer learning method between two datasets with different MI tasks. The proposed pre-training method requires only minimal fine-tuning data when applying DL models to new MI paradigms, making MI-Brain–computer interface more practical and user-friendly.

The experimental and theoretical spectroscopic elucidation of molecular structure, electronic properties, thermal analysis, biological evaluation, and molecular docking studies of isococculidine

Experimental techniques, such as vibrational spectroscopy (FT-IR and FT-Raman), UV–visible, thermal analysis (DSC), along with quantum chemical calculations based on density functional theory (DFT) were conducted to explore the physicochemical properties of a plant-derived natural product, isococculidine. The potential energy distribution (PED) has been assigned to each of the vibrational bands observed in FT-IR and FT-Raman spectra. The reduced density gradient (RDG) analysis was used to demonstrate the non-covalent interactions (NCI). Natural bond orbital (NBO) study revealed that the benzene ring's hyper-conjugative interactions were primarily responsible for the molecular stability. The intramolecular charge transfer has been further verified by the HOMO-LUMO analysis. The molecular electrostatic potential (MEP) analysis signifies that the nucleophilic and electrophilic interaction sites are prone to oxygen of O1-CH3 and hydrogens of O2-CH3, respectively. The chemical activity and the toxicity of the molecule have been examined based on the calculated Mulliken charges, softness, hardness, and local reactivity parameters. The thermal properties like specific heat capacity, entropy and enthalpy which affect the drug's activities were studied. The large value of the first hyperpolarizability indicates its significant nonlinear optical (NLO) property. The molecular docking of isococculidine has been performed with nicotinic acetylcholine protein which deliberates the significant biological activity of the molecule.

Deep multitask learning with label interdependency distillation for multicriteria street-level image classification

Multitask learning (MTL) aims at beneficial joint solving of multiple prediction problems by sharing information across different tasks. However, without adequate consideration of interdependencies, MTL models are prone to miss valuable information. In this paper, we introduce a novel deep MTL architecture that specifically encodes cross-task interdependencies within the setting of multiple image classification problems. Based on task-wise interim class label probability predictions by an intermediately supervised hard parameter sharing convolutional neural network, interdependencies are inferred in two ways: i) by directly stacking label probability sequences to the image feature vector (i.e., multitask stacking), and ii) by passing probability sequences to gated recurrent unit-based recurrent neural networks to explicitly learn cross-task interdependency representations and stacking those to the image feature vector (i.e., interdependency representation learning). The proposed MTL architecture is applied as a tool for generic multi-criteria building characterization using street-level imagery related to risk assessments toward multiple natural hazards. Experimental results for classifying buildings according to five vulnerability-related target variables (i.e., five learning tasks), namely height, lateral load-resisting system material, seismic building structural type, roof shape, and block position are obtained for the Chilean capital Santiago de Chile. Our MTL methods with cross-task label interdependency modeling consistently outperform single task learning (STL) and classical hard parameter sharing MTL alike. Even when starting already from high classification accuracy levels, estimated generalization capabilities can be further improved by considerable margins of accumulated task-specific residuals beyond +6% κ. Thereby, the combination of multitask stacking and interdependency representation learning attains the highest accuracy estimates for the addressed task and data setting (up to cross-task accuracy mean values of 88.43% overall accuracy and 84.49% κ). From an efficiency perspective, the proposed MTL methods turn out to be substantially favorable compared to STL in terms of training time consumption.