Hajee Mohammad Danesh Science and Technology University

The nutritional enhancement and optimum physical properties in formulating instant fruit juice powder are challenging. This study explored the development and evaluation of freeze‐dried instant fruit juice powders from three tropical fruits: Guava (Psidium guajava), Amla (Phyllanthus emblica), and Jamun/black plum (Syzygium cumini), aimed at offering health‐promoting benefits and ensuring year‐round supply. The freeze‐drying method was chosen so that it helps in retaining the nutritional and sensory qualities of the fruits. Consequently, three types of juice powders were developed: Guava Juice Powder (GJP), Amla Juice Powder (AJP), and Jamun Juice Powder (JJP), through the mixing of dried fruit powders with other key ingredients. These juice powders were evaluated for their physical, chemical, and functional characteristics, as well as consumer appeal. The findings revealed that the powders successfully preserved the physical qualities essential for good storage and consumer use. They retained high levels of vitamin C (116.21–176.89 μM), carotenoids (5.09–8.03 μM β‐carotene), phenolics (289.56–822.62 mg GAE/100 g), flavonoids (98.21–607.74 mg QE/100 g), and antioxidant activity (43.05–45.90 μM Trolox equivalents), highlighting their potential as functional food ingredients. Sensory evaluations showed high acceptability, particularly for JJP (8.29), with GJP (7.97) and AJP (7.74) also performing well. This research contributes to the functional foods market by offering a novel method for utilizing tropical fruits, thereby expanding consumer options, supporting health, and minimizing fruit waste.

Reaction of copper(ii) acetate with (E)-2-(((2-benzoylphenyl)imino)methyl)phenol (HL1), (E)-2-(((2-benzoyl-5-chlorophenyl)imino)methyl)phenol (HL2) and (E)-1-(((2-benzoyl-5-chlorophenyl)imino)methyl)-naphthalen-2-ol (HL3) provided bis[(E)-2-(((2-benzoylphenyl)imino)methyl)phenolato-κ²N,O]copper(ii) (1), bis[(E)-2-(((2-benzoyl-5-chlorophenyl)imino)methyl)phenolato-κ²N,O]copper(ii) (2) and bis[(E)-1-(((2-benzoyl-5-chlorophenyl)imino)methyl)naphthalen-2-olato-κ²N,O]copper(ii) (3). The molecular structure determinations revealed that the ligands existed as a usual (imine)N⋯H–O(phenol) (enolimine-form) in the solid state, which was further evidenced using ¹H NMR studies in solution (CDCl3 and DMSO-d6). Unlike HL1 and HL2, two symmetry-independent molecules (A and B) were present in the unit cell of the HL3 crystal. The molecular structures showed that the two N,O-chelating ligands coordinate to the copper(ii) ion through a square-planar (1), a distorted square-planar (2) and a square-pyramidal geometry (3). Each asymmetric unit of the crystal structure contained one-half of the molecule for 1, a single molecule for 2 and two symmetry-independent molecules for 3 (molecules A and B). Thermal investigations using DSC demonstrated an irreversible phase transition from a crystalline solid to an isotropic liquid (m.p.). Cyclic voltammogram results proved two quasi-reversible one-electron charge transfer process for 1 and 3 in DMF at 25 °C. Complexes 1 and 2 exhibited low and significant antibacterial activity, respectively, against E. coli and S. aureus, while 3 was completely inactive. Among the ligands, only HL2 exhibited medium activity against microorganisms. The electronic and molecular structures correlated well with the computational modeling performed using DFT/TD-DFT calculations.

Accurate forecasting of bike-sharing demand is essential for optimizing fleet management and improving user satisfaction. Traditional models often struggle to capture the complex spatiotemporal dependencies required for high-quality predictions. This study proposes the ApexBoost Regression (ABR) model, which integrates decision tree regression with gradient-based boosting to enhance the modeling of intricate temporal and spatial patterns in bike rental behavior. The model was applied to the Metro Bike Sharing dataset, where the data were partitioned chronologically: the full year of 2023 (36,312 data points, approximately 80%) was used for training, and the fourth quarter of 2024 (9,118 data points, approximately 20%) was reserved for testing. To improve predictive performance, we conducted extensive feature extraction, including spatial variables (longitude, latitude, and location cluster), temporal indicators (hour, day of week, day of month, is_morning, is_evening), and cyclical encodings (sine and cosine transformations of hour). We then employed a feature selection algorithm (FSA), guided by both correlation analysis and feature importance metrics, to retain the most informative and non-redundant variables. Our exploratory data analysis (EDA) revealed important usage patterns such as peak rental times across seasons and day types, supporting feature selection choices.Comparative evaluation against state-of-the-art models-including XGBoost and Random Forest-demonstrates the superior performance of ABR. Prior to hyperparameter tuning, the model achieved a mean squared error (MSE) of 3.0501. After optimization using particle swarm optimization (PSO), the MSE was further reduced to 3.0164, confirming the effectiveness of both the proposed model and the optimization strategy.

As industrialization and the development of smart cities progress, effective waste collection, classification, and management have become increasingly vital. Recycling processes depend on accurately identifying and restoring waste materials to their original states, essential for reducing pollution and promoting environmental sustainability. In recent years, deep learning (DL) techniques have been applied strategically to enhance waste management processes, including capturing, classifying, composting, and disposing of waste. In light of the current context, the study presents an innovative waste classification model that utilizes a tailored DenseNet201 architecture coupled with an integrated Squeeze and Excitation (SE) attention mechanism and the fusion of parallel Convolutional Neural Network (CNN) branches. The integration of SE attention enables squeezing the irrelevant features and excites the important ones and the fusion of parallel CNN branches enhances the extraction of intricate, deeper, and more distinguishable features from waste data. The evaluation of the model across four publicly available datasets, along with three additional datasets to enhance waste diversity and the model’s reliability, and the incorporation of Grad-CAM to visualize and interpret the model’s focus areas for transparent decision-making, confirms its effectiveness in improving waste management practices. Furthermore, this model’s successful deployment in a web-based sorting system marks a tangible stride in translating theoretical advancements into on-the-ground implementation, promising heightened efficiency and scalability in waste management practices. This work presents a precise solution for adaptable waste classification, heralding a paradigm shift in global waste disposal norms.

As human capital forms the backbone of any organization, managing and minimizing employee attrition is of paramount importance. Attrition prediction is essential because attrition disrupts projects advancement, increases rehiring and training costs, and risks losing core knowledge and technologies. Thus, minimizing attrition is essential for organizational stability and competitiveness. In this research, we design a Machine Learning (ML)-based predictive system using six ML models namely Logistic Regression, Random Forest, Gradient Boosting (GB), Decision Tree, Support Vector Machine, and K-Nearest Neighbors and evaluate their performances using accuracy, precision, recall, and f1 score. This system predicts employee attrition in advance, allowing organizations to implement proactive talent management strategies instead of relying on traditional reactive approaches. We analyse ‘IBM HR Analytics data’, that contains 1,470 observations and 35 features. Among all the models, GB outperform others with 98% accuracy, and precision, 100% recall and 99% f1 score. Furthermore, using Explainable AI (XAI) methods such as ELI5, SHAP, LIME, and SHAPASH, we analyze the most influencing factors related to employee attrition. XAI analysis show that ‘Over Time’ ‘Toal working years’, ‘job level’ are the most significant factors. Local explainability provide the low level explanation to reveal the inner story and deep insight.

This paper presents 3D modeling for lightning electromagnetic effects to analyze the peak induced voltages on overhead conductors using the finite element method (FEM) and COMSOL Multiphysics software. It mainly focuses on the parameters that affect the peak induced voltage of overhead conductors, such as the height of the conductor from the ground, the conductivity of the ground, and the speed of the source current. Initially, a thorough observation of the return stroke current for lightning strikes to flat ground at different heights along the lightning channel and induced voltage at the center point of the horizontal conductor at distances of 40 m, 60 m, and 100 m from the stroke point has been carried out for model validation. Then the effect of conductor height on the peak induced voltages on overhead conductors, with specific numerical results (e.g., peak induced voltage reaching 75.70 kV at the center point and − 14.87 kV at the terminal point for the 50 m horizontal distance from the stroke point, 4.8 m lower conductor height, and finite ground). It observed that there is a significant impact on peak induced voltages at the center and terminal points at different conductor heights for finite ground conductivity. The findings also show that for finite ground conductivity (0.001 Sm−1), the peak induced voltage at the center point is more than twice the value observed in a perfectly electric conducting ground (5.98 × 107 Sm−1) at a 4.8 m lower conductor height.

Agroforestry, the integration of trees with crops and/or livestock, plays a vital role in promoting socioeconomic stability and environmental sustainability, particularly in arid and forest-deficient regions. Despite its documented benefits, adoption remains limited due to various socioeconomic and perceptual barriers. This study employed a structured questionnaire survey involving 250 randomly selected farmers across five tehsils of Bahawalpur district. Data were analyzed using SPSS, including chi-square tests and mean rank scores, to assess perceptions, practices, and economic impacts of agroforestry systems. The majority of respondents preferred monoculture cropping (71.35%) over agroforestry (28.65%). The main agroforestry systems identified were Agri-Silviculture (47.6%), Agrisilvopastoral (40%), and Silvopastoral (12.4%). Multipurpose tree species such as Vachellia nilotica, Albizia lebbeck, and Dalbergia sissoo contributed up to PKR 25,000/acre annually. A significant number of respondents (30%) relied on fuelwood, valued up to PKR 15,000. Tree density ranged from 5–12 trees per hectare, with a potential increase to 40 trees/hectare. Despite low adoption, statistical analysis revealed a significant positive impact (P < 0.05) of agroforestry on household income and livelihood resilience during economic and climatic shocks. Findings reveal low agroforestry adoption due to perceived crop yield losses; however, practitioners report financial benefits highlighting agroforestry’s untapped potential. Satellite data show minimal forest cover (<1%), reinforcing the need for tree integration. Agroforestry holds strong potential for enhancing rural livelihoods and combating environmental degradation in arid regions. However, misperceptions regarding reduced crop yields hinder wider adoption. Strategic extension services, awareness programs, and financial incentives are critical to encourage agroforestry adoption and maximize its socioeconomic and environmental benefits.

Deep learning has revolutionized the classification of land cover objects in hyperspectral images (HSIs), particularly by managing the complex 3D cube structure inherent in HSI data. Despite these advances, challenges such as data redundancy, computational costs, insufficient sample sizes, and the curse of dimensionality persist. Traditional 2D Convolutional Neural Networks (CNNs) struggle to fully leverage the interconnections between spectral bands in HSIs, while 3D CNNs, which capture both spatial and spectral features, require more sophisticated design. To address these issues, we propose a novel multilayered, multi-branched 2D-3D CNN model in this paper that integrates Segmented Principal Component Analysis (SPCA) and the minimum-Redundancy-Maximum-Relevance (mRMR) technique. This approach explores the local structure of the data and ranks features by significance. Our approach then hierarchically processes these features: the shallow branch handles the least significant features, the deep branch processes the most critical features, and the mid branch deals with the remaining features. Experimental results demonstrate that our proposed method outperforms most of the state-of-the-art techniques on the Salinas Scene, University of Pavia, and Indian Pines hyperspectral image datasets achieving 100%, 99.94%, and 99.12% Overall Accuracy respectively.

Floods are among the most devastating natural disasters, causing extensive property damage and loss of life. An effective flood early warning system can significantly reduce such damage and save lives. However, data related to flood characteristics are often non-linear and uncertain, which leads to poor predictive accuracy in conventional hydrological models. Traditional models struggle to handle these complexities due to their rigid assumptions, limited adaptability to regional variability, susceptibility to input uncertainties, and inability to capture abrupt changes caused by extreme weather events. This study aims to enhance the prediction of flood occurrence and danger levels in non-tidal rivers by proposing a hybrid deep learning (DL) model that integrates a Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM) network, and an Attention mechanism—collectively referred to as the CNN-LSTM-Attention model. Furthermore, we incorporate Explainable AI (XAI) tool SHAPASH, to interpret the model’s predictions. The proposed model is benchmarked against several state-of-the-art DL algorithms using a dataset that includes water level, rainfall, discharge, and other relevant hydrological variables. Experimental results demonstrate that the CNN-LSTM-Attention model outperforms all baseline models in predicting flood occurrence and water danger levels. It achieves 98.63% accuracy with 99% precision, recall, and F1 score for water danger level prediction, and 97.26% accuracy with 95% precision, 97% recall, and 96% F1 score for flood prediction. From a global explainability perspective, SHAPASH indicates that water level is the most influential feature in flood prediction, followed by discharge and rainfall. Local explainability results show that predictions are primarily driven by key features, with lesser influential variables contributing marginally. This study provides a robust and interpretable flood prediction framework, offering valuable insights for hydrologists and climate scientists.

Mangoes are highly nutritious fruits with significant commercial value; however, they are prone to postharvest diseases and short shelf life, leading to substantial losses. This paper addresses concerns regarding chemical residues without causing any economic losses through the development of an edible coating with chitosan and turmeric extract to prolong postharvest shelf life and preserve mango quality. The experiment followed a Completely Randomized Design (CRD) with three replications, each containing ten fruits. The five treatments were: Control (T0), 1.5% chitosan (CHT) (T1), 1.5% chitosan + 0.5% turmeric extract (TUE) (T2), 1.5% chitosan + 1% turmeric extract (T3), and 1.5% chitosan + 1.5% turmeric extract (T4). Physiologically mature mangoes were treated with these coatings, and quality parameters were evaluated during storage. Results showed reduced weight loss and better fruit firmness in coated mangoes. Furthermore, the coatings also influenced total soluble solids, titratable acidity, vitamin C, sugar, and Beta-carotene content compared to the control. Disease incidence was also considerably reduced by the coating treatments. Treatment with 1.5% chitosan + 1% turmeric extract produced the best results for total phenol, flavonoid, and antioxidant (DPPH* scavenging activity) over the control. Treated mangoes retained better color parameters (L*-Lightness, a*-greenness to redness (green to red transition), b*-yellowness), preserving visual appeal. Sensory evaluations revealed improved flavor, sweetness, appearance, and overall impression in treated mangoes. Overall, combining chitosan and turmeric extract effectively reduced postharvest losses and enhanced mango shelf life and quality, offering a safe, eco-friendly alternative to chemical treatments.

Background and Aims Myopia is a prevalent refractive error, particularly among young adults, and is becoming a growing global concern. This study aims to predict myopia among undergraduate students using ensemble machine learning techniques and to identify key risk factors associated with its development. Methods A cross‐sectional study was conducted in Dinajpur city, collecting 514 samples through a self‐structured questionnaire covering demographic information, myopia prevalence and risk factors, knowledge and attitudes, and daily activities. Four feature selection techniques Boruta‐based feature selection (BFS), Least Absolute Shrinkage and Selection Operator regression, Forward and Backward Selection and Random Forest (RF) identified 12 key predictive features. Using these features, ensemble methods, including logistic regression artificial neural network, RF, Support Vector Machine, extreme gradient boosting, and light gradient boosting machine were employed for prediction. Model performance was evaluated using accuracy, precision, recall, F1‐score, and area under the curve (AUC). Results The stacking ensemble model achieved the highest performance, with an accuracy of 95.42%, recall of 93.42%, precision of 98.85%, F1‐score of 96.08%, and AUC of 0.979. SHapley Additive exPlanations analysis identified key risk factors, including visual impairment, family history of myopia, excessive screen time, and insufficient outdoor activities. Conclusion These findings demonstrate the effectiveness of ensemble machine learning in predicting myopia and highlight the potential for early intervention strategies. By identifying high‐risk individuals, targeted awareness programs and lifestyle modifications can help mitigate myopia progression among undergraduate students.

Accurate and reliable plant disease detection systems have to be implemented for optimizing agricultural profitability and preserving global food security. Deep learning-based image classification has shown significant potential in addressing these challenges. However, its application in resource-limited demands of ecological responsibility solutions that are fast, accurate, and computationally efficient. In order to satisfy these requirements, we have employed a compact transfer learning architecture for rice leaf disease identification that uses nine disease classes. The framework features preprocessing approaches such as illumination correction to enhance image quality. The method we propose integrates a tailored classifier network with a pretrained EfficientNetB0 model for optimal feature extraction and precise classification. With the objective to ensure its efficacy, we evaluate the proposed framework against multiple transfer learning models, including DenseNet121, VGG16, InceptionV3, MobileNetV2, and ResNet50. In accordance with the experimental results, the proposed customized EfficientNetB0 framework exceeds the other frameworks, attaining 98.16% training accuracy and 94.47% testing accuracy. In comparison, DenseNet121, VGG16, InceptionV3, MobileNetV2, and ResNet50 achieved training accuracies of 77.67%, 75.59%, 91.74%, 78.94%, and 93.16%, and testing accuracies of 71.75%, 71.58%, 89.02%, 77.28%, and 89.52%, respectively.

Biosensors are essential tools for detecting and analyzing various elements of human biology. This study introduces an innovative circular-shaped photonic crystal fiber (PCF) with a hexahedron core for the precise detection of blood components. The sensor’s performance evaluated using COMSOL Multiphysics software. The finite element methods (FEM) is applied to solve Maxwell’s equations and perform simulations across a terahertz (THz) frequency range from 1.0 to 3.0 THz. This comprehensive investigation focuses on optimizing several important optical parameters, including relative sensitivity (RS), confinement loss (CL), effective mode area (EMA), and birefringence, etc. for enhancing the detection of various blood components. The Optical sensor is constructed by Topas as cladding material. The sensor demonstrates exceptional performance with RS of approximately 95.02% for glucose, 95.48% for plasma, 96.30% for white blood cells (WBCs), and 97.04% for red blood cells (RBCs) at an operational frequency of 2.20 THz. Thus the proposed sensor can provide reliable and accurate measurements across different blood components in advanced biomedical applications.

Freeze drying and pretreatment have potential impact on nutritional quality of vegetables. This study was sought to know the effect of pretreatment followed by freeze and hot air drying on quality of dehydrated mushroom powders. Blanching as pretreatment was done through hot water, potassium metabisulfite (KMS) and steam blanching. Physicochemical (composition, physical properties and color), minerals (calcium, potassium, iron, manganese and zinc), biochemical properties (vitamin C, total phenol content, and flavonoids, ergothioneine and antioxidant activity), functional groups and micro-structure were evaluated for the dehydrated product. Ash, protein, dietary fiber, fat and carbohydrate of the dehydrated mushroom varied between 7.13–12.48%, 3.51–5.44%, 9.72–12.60%, 1.22–1.85%, and 70.11–78.98%, respectively. Calcium, potassium, iron, manganese and zinc ranged between 0.02–0.84%, 0.91–3.99%, 59.1–384.6 ppm, 6.1–16.1 ppm and 106–265 ppm, respectively. Freeze dried sample possessed the highest value of L* and the lowest value of a* and b*. Total phenol content, ascorbic acid, flavonoids, ergothioneine and antioxidant activity of the dehydrated mushroom powder outlined between 0.57–4.03 mg/100 g, 0.14–0.49 mg/100 g, 14.55–35.45 mg/100 g, 1.65–2.34 mg/g and 32.1–85.54%, respectively. Quality degradation of the dehydrated product was lower for steam blanching than hot water and KMS blanching. Scanning electron microscopic analysis showed the uniform pore space for freeze drying.

The present dissertation reports a comprehensive investigation of the structural, bulk, electronic, optical and thermodynamics features of Th-based intermetallics compounds ThX2Si2 (X = Ru, Rh, Ir, Pt). All the investigation was completed using ab initio scheme depend upon the density functional theory. Extremely well concurrence involving the experimental records and our calculated values of all the compounds has been observed. The dynamical and structural stability is ensured from phonon dispersion curves and formation enthalpy calculations. The investigated elastic constants show positive values and hold Born’s stability criteria which confirmed the mechanical stable nature of these materials. The large bulk and Young’s moduli of ThX2Si2 (X = Ru, Rh, Ir, Pt) ensured their high stiffness characteristics whereas ThIr2Si2 carries high stiffness characteristics. Except the phase ThIr2Si2, the phases ThRu2Si2, ThRh2Si2, ThPt2Si2 show ductility nature as ensured from Pugh's ratio and Poisson's ratio data. According to hardness calculations only the phase ThIr2Si2 shows hard nature whereas the phase ThRh2Si2 lies on the hardness border line. On the other hand, the remaining phases ThRu2Si2 and ThPt2Si2 show soft nature as they have hardness value below 10 GPa. High machinable index, µm of ThPt2Si2 compared to other phases ensures it high industrial application for cutting tool geometry, cutting fluids and so on. ThX2Si2 (X = Ru, Rh, Ir, Pt) materials exhibit metallic behavior ensured by the explanation of band structure, DOS and optical phenomena. At UV energy region, the major peak of absorption and conductivity is observed. The studied intermetallics can be good candidates for solar reflector in UV region as they possess high reflectivity in UV zone. ThRh2Si2 is thermally more conductive among other three compounds because of its large Debye temperature. The high melting temperature and extremely lower thermal conductivity of ThX2Si2 (X = Ru, Rh, Ir, Pt) ensured their potential use in Thermal Barrier Coating (TBC) materials.

This paper presents a new cancer detection sensor using an octagonal photonic crystal fiber (PCF) to identify cancerous tissues in the breast, cervix, and skin. The sensor is designed to provide highly accurate detection due to the advanced properties of PCF. The study uses the finite element method (FEM) and MATLAB to design and analyze the sensor. The results show a strong mathematical evaluation of its performance across the 1.0–3.0 THz frequency range. Notably, the sensor achieves a relative sensitivity of approximately 97% and a confinement loss of about 10−8 dB/m at 2.2 THz for all investigated breast, cervical, and skin. Furthermore, the lowest effective material loss for breast is 0.0047155 cm−1 at 2.2 THz. This sensor uses the unique photonic properties of cancer cells to quickly and accurately detect breast, cervical, and skin cancers. Its small size and flexible design allow for minimally invasive procedures, making it suitable for real-time cancer diagnosis in biomedical applications.

Background Polystyrene microplastics (PS‐MPs) are pervasive pollutants impacting animals across ecosystems, including livestock and wildlife, through contaminated food, water, and air. MPs may disrupt endocrine function, particularly affecting the thyroid gland, which is essential for metabolism and development. Objectives This study investigates the effects of PS‐MPs on thyroid function in mice, offering insights relevant to veterinary care by examining changes in gene expression and biochemical markers. Methods PS‐MPs of 5 µm diameter were prepared in distilled water after probe sonication. Sixty male Swiss albino mice were divided into three groups: a control group and two treatment groups receiving 0.1 mg and 0.2 mg PS‐MPs via oral gavage for 28 days. Mice were anesthetised, and thyroid tissues were collected for histopathological, biochemical, and gene expression analyses. Biochemical tests included catalase, superoxide dismutase, reactive oxygen species, and hormone levels. Histopathology and gene expression (TSHR and TPO) of thyroid‐related genes were examined to assess PS‐MPs induced effects. Results Exposure to PS‐MPs in mice led to significant increases in calcium, thyroxin, free T3, free T4, ALP, AST, ALT, and amylase levels, alongside elevated oxidative stress markers. Conversely, the levels of TSH, calcitonin, magnesium and phosphate decreased. Histopathological analysis showed abnormal thyroid follicle development, decrease parafollicular cells, with colloid loss, haemorrhage, and necrosis. Gene expression analysis revealed a marked reduction in TSHR and TPO levels in PS‐MPs treated groups, indicating thyroid dysfunction. These findings highlight the profound impact of PS‐MPs on thyroid gland function in mice. Conclusion These findings underscore the potential risks that PS‐MPs pose to thyroid health, with potential consequences for other veterinary species. As environmental contamination rises, veterinarians may encounter more endocrine disorders linked to PS‐MPs, emphasising the need for further research and preventive measures.