Weedy rice (Oryza sativa f. spontanea) poses significant challenges to rice production, reducing yield and commercial value. This study investigates the inhibitory effects of root, stem, and leaf extracts from cultivated rice OM5930 (60 days after sowing) on two weedy rice lines, WR19 (short awn) and WR20 (long awn). The objectives were to: (i) identify the most potent plant part for inhibition, (ii) determine the optimal treatment duration, and (iii) assess the resistance of the two weedy rice lines. Results demonstrated that extracts from all OM5930 plant parts suppressed seedling shoot and root growth in both weedy rice lines. Leaf extracts exhibited the strongest inhibition, achieving complete suppression (100%) at 0.3 g/mL across all time points (0, 48, and 96 hours). The 48-hour treatment showed the most stable inhibitory effect. Root and stem extracts displayed lower efficacy, reaching only 60–70% inhibition at the same concentration. WR20 was more susceptible than WR19, with a slight stimulation observed at lower concentrations (0.015 and 0.075 g/mL). The strong allelopathic potential of OM5930 leaf extracts suggests their application as an eco-friendly bioherbicide, reducing dependence on synthetic herbicides, minimizing production costs, and promoting sustainable rice cultivation.

The study investigated the impact of probiotics and vitamins (PV) on the reproductive performance and egg quality of Ac chickens' diet. The experiment was conducted using a completely random design with 250 hens, aged 40-50 weeks, housed in cages. There were 5 treatments, each with 5 repetitions, and 10 hens per repetition. The treatments corresponding to the diets were: Control (BD), C250 (BD + 250 mg PV/kg of feed), C500 (BD + 500 mg PV/kg of feed), C750 (BD + 750 mg PV/kg of feed), and C1000 (BD + 1000 mg PV/kg of feed). The results showed that the highest laying rate (39.2%), yolk index (0.45), albumen index (0.08) and yolk colour (7.6) were in C1000. Although the C750 indicated the lowest FCR (3.78), it had the greatest egg weight (36.4 g/egg) and Haugh unit (82.8). In conclusion, probiotics and vitamins at 1,000 mg/kg diet improved Ac chicken laying rate, FCR, and egg weight.

Diverse starch-based foods have been widely consumed. To keep the products in good quality, preservatives added are inevitable. Sodium benzoate is one of the most commonly used food preservatives for starch-based products. As an antimicrobial compound, high concentration of sodium benzoate present in water can pose risks to aquatic indigenous microbial communities. In this study, 27 sodium benzoate degrading bacteria were isolated from wastewater samples collected at three rice noodle production facilities. Isolates SB2.1, SB2.2, SB2.4, SB3.10, and SB3.13 demonstrated effective degradation of sodium benzoate at concentrations of 1250, 2500, and 5000 mg/L. Optimal degradation efficiency (>93%) at 1250 mg/L was observed when the isolates were grown in MM medium at pH ranging from 7 to 9. Notably, SB3.10 exhibited chemotaxis towards sodium benzoate after 24 hours of incubation. Based on 16S rRNA gene sequencing and biochemical characterization, SB3.10 was identified as Acinetobacter calcoaceticus SB3.10.

Human-action recognition aims to identify the actions performed by individuals. Due to the broad spectrum of human activities, action recognition covers a wide range. Among all, fall detection is a critical aspect of surveillance, particularly in environments where individuals are at risk. Throughout the years, several sensors, data types, and classification techniques have been investigated to address this issue. This paper proposes a lightweight fall detection designed to process sequences of images in real-time. This system is deployed on an embedded device, specifically the Jetson nano. Our goal is to construct a comprehensive dataset that accurately detects falls in various lighting conditions. The proposed system is constructed using YOLOv8n-pose, which have been trained to identify people using widely recognized dataset. Our methodology includes the design and implementation of the YOLOv8n-pose, data collection, and rigorous testing to ensure the accuracy of fall detection in real-time using surveillance camera. The experimental results show that high detection accuracy and acceptable timing capabilities are achieved.

This study presents a comparative analysis of state-of-the-art deep learning models–EfficientNetB0, MobileNetV2, and ResNet101–for image classification and content-based retrieval in scientific publications. A dataset of 4,303 images from 11 categories was curated from the Can Tho University Journal of Science and enhanced through tailored data augmentation strategies. The models were fine-tuned using transfer learning with hyperparameters optimized via Grid Search. Features were extracted using GlobalAveragePooling2D, and cosine similarity combined with the FAISS library was employed for efficient similarity search. Experimental results demonstrate a clear performance-efficiency trade-off: ResNet101 achieved the highest classification accuracy, while EfficientNetB0 and MobileNetV2 offered significant advantages in inference speed. A user-friendly web interface was developed to support practical image retrieval applications. These findings highlight the potential of deep learning in enhancing the management and integrity of scientific image resources.

This literature review synthesizes recent advancements and emerging trends in the field of project management, with a focus on the evolving impact of technology, leadership, and sustainability. Drawing on a wide range of scholarly articles and industry publications, the review explores the foundational principles of project management, such as risk management, leadership effectiveness, and project efficiency, while also highlighting the complexities of applying these practices across various sectors. The integration of cutting-edge technologies like artificial intelligence, data analytics, and hybrid methodologies is identified as a key driver of change in project execution. Furthermore, the increasing emphasis on sustainability and ethics reflects a shift in the industry towards more comprehensive measures of project success, beyond traditional financial outcomes. This review emphasizes the need for continuous adaptation in project management practices and calls for future research to refine leadership strategies, improve technology adoption, and develop more holistic frameworks for assessing project success. The findings underscore the importance of aligning traditional project management principles with the demands of an increasingly complex and fast-evolving global environment.

This paper studied the diversity of phytoplankton species at 9 sampling points of aquaculture ponds in 3 districts of Dak Ha, Sa Thay, Kon Plong, Kon Tum province in 2022  ̶  2023. The results recorded 7 phyla, 98 genera and 142 species, of which 22 genera and 23 species of phytoplankton appeared in all 3 sampling periods. The values ​​of the biodiversity index (H’) and the diversity value index (Dv) at 9 sampling points ranged from 2.51 - 3.66 and 2.39 - 3.58, respectively, indicating that the diversity of phytoplankton in aquaculture ponds is high and very high. The regulation index (J) ranged from 0.95 - 0.99, indicating that phytoplankton species were evenly distributed at all studied sampling points. The similarity coefficient of species composition of phytoplankton populations at 9 sampling points ranged from 0.43 - 0.85. The number of phytoplankton species in water samples collected at 9 studied locations had a negative correlation with the impedance value of the water samples. In particular, the main environmental factors affecting the cell density of phytoplankton communities at the sampling locations in this study were suspended solids, salinity, and conductivity, which had a positive correlation, and a negative correlation with impedance.

This study probed into how gender and development (GAD) concepts and principles are included in English 10 learning materials through a mixed-methods approach. The examination of lesson plans and self-learning modules through content analysis revealed biases supporting traditional gender roles, such as inadequate representation of women and showing stereotypical gender behaviors. Interviews with English 10 public teachers highlighted diverse strategies employed by teachers, despite limited exposure to GAD concepts, indicating a lack of comprehensive training in this area.; However, teachers still endeavor to integrate these principles into their instruction to foster inclusivity, though challenges persist, such as biases influencing disciplinary measures. The findings of this study highlight the importance of challenging narrow views and promoting inclusive practices in educational materials. This study contributes to the ongoing efforts to promote gender equality and diversity in education highlighted in UNESCOs Education for All initiative (2003), emphasizing the need for continuous awareness, sensitivity, and proactive integration of GAD principles in educational settings. Keywords: content analysis, gender and development, learning materials, inclusivity in education, mixed-methods, SDG 4

This study explores the factors influencing online learning motivation among students at Can Tho University. The research examines five key factors: personal, lecturer-related, institutional, academic, and environmental aspects. A survey was conducted with 892 students across various academic disciplines. The study employed statistical analyses, including Cronbach’s Alpha reliability testing, exploratory factor analysis (EFA), and multiple regression analysis to determine the impact of these factors on students' motivation.The findings indicate that personal and lecturer-related factors have the most significant positive influence, highlighting the importance of self-discipline, time management, and lecturer support. Institutional factors also play a crucial role, particularly in terms of learning infrastructure and support services. However, environmental factors negatively affect motivation, as poor internet connectivity, financial difficulties, and distractions hinder students’ engagement.  Additionally, students with higher academic performance and greater online learning experience show stronger motivation. Differences across academic disciplines suggest the need for tailored teaching methods and institutional support. These findings provide insights into enhancing online learning motivation and contribute to policy recommendations for improving the quality of online education at Can Tho University.

The aim of this study was to investigate the antimicrobial activity against Staphylococcus aureus, Propionibacterium acnes, Escherichia coli, and Candida albicans of the ethanol extract and the n-hexane and ethyl acetate extract of the stingless bee Tetragonula sp. propolis. Additionally, the study also identified the stingless bee species based on morphological characteristics and COI gene sequence. The antimicrobial activity of three stingless bee propolis extracts showed resistance against all three bacterial S. aureus, P. acnes, and E. coli. At a concentration of 1000 μg/mL, the ethanol extract exhibited the highest resistance, with P. acnes showing a zone of inhibition of 13.33 mm, while the inhibition zones were 11.33 mm for S. aureus and 10.33 mm for E. coli. The Minimum Inhibitory Concentration (MIC) for the three bacterial S. aureus, P. acnes, and E. coli were found to be 5, 10, and 15 μg/mL, respectively. The study identified the presence of polyphenols, flavonoids, aglycones, phenolics, and ketones  quantified the total phenolic and flavonoid content, assessed the antibacterial activity of the propolis extract against the three pathogenic bacteria, and provide the foundation for further research.

The application of machine learning models in the analysis of helmet-related images has yielded remarkable results in identifying and classifying helmet wearing behaviors. Previous research has used several pretrained models to predict proper or improper helmet use, achieving the highest accuracy of 98.61% on the Helmet Wearing Image Dataset (2024), a newly introduced dataset designed to improve the ability to classify helmet wearing behaviors. This study aims to improve the prediction performance on helmet datasets by leveraging state-of-the-art deep learning models combined with ensemble techniques. Using ResNet-50, MobileNetV2, and EfficientNet-B0 models, the proposed EnsemHelmet Framework uses soft voting ensemble to optimize the classification results, achieving an outstanding accuracy of 99.18% on the experimental dataset. The results demonstrate the potential of ensemble learning to achieve high performance. This study not only improves the accuracy of the helmet wearing recognition system but also highlights the effectiveness of ensemble techniques in optimizing performance on real-world datasets.