Adenosma bracteosum Bonati has been utilized in traditional and modern Vietnamese medicine for the treatment of hepatitis, lung, and liver diseases. Bacteria that reside within the cells of medicinal plants, utilizing unique strategies to enhance the growth and survival of their host plants, often through distinctive secondary metabolites, are known as symbiotic or endophytic bacteria. In this study, 58 endophytic isolates were obtained from the wild medicinal plant Adenosma bracteosum Bonati and were assessed for their in vitro antibacterial activities against common pathogenic bacteria including Escherichia coli, Staphylococcus aureus, Aeromonas hydrophila, Vibrio parahaemolyticus, and Dickeya dadantii. Twelve isolates with broad antibacterial activity produced siderophores and lytic enzymes, with SB1R13.2 showing the greatest resistance against all five pathogenic bacterial strains, producing siderophores and synthesizing digestive enzymes. According to the 16S rDNA sequences, the SB1R13.2, SB4R5, and SB5T2 isolates demonstrated the most similar genetic affinity to Bacillus velezensis. Meanwhile, the SB4R2 isolate exhibits genetic similarity to Burkholderia sp. These findings suggest that this specific species could be a promising antibacterial agent, potentially offering resistance against other pathogens

This study examines the factors limiting research output in Vietnam, particularly the publication of Web of Science (WoS) listed papers, which affects the global ranking of Vietnamese universities. We analyzed the academic profiles of 2,042 faculty members from the top 12 universities, as ranked by Webometrics in 2021, using a nonlinear Poisson regression model. The analysis identified eight key factors influencing publication count: age, gender, field of study, academic degree, academic rank, professional title, administrative position, and educational background. Based on these findings, we propose recommendations to boost international publication productivity in Vietnamese universities.

This study investigates the impact of inner-loop pressure regulation on the dynamic performance of pneumatic artificial muscle (PAM) systems using a dual-loop control architecture. Three pressure control strategies – Proportional-Integral (PI), Proportional-Integral-Derivative (PID), and Radial Basis Function neural network-tuned PID (RBF-PID) – are experimentally evaluated in terms of tracking accuracy, transient response, and disturbance rejection. Results show that the RBF-PID controller achieves the highest pressure tracking accuracy, with a root mean square error (RMSE) of 0.067 bar under a modulated sinusoidal input, outperforming PID (0.088 bar) and PI (0.094 bar) controllers. In position control tasks, all dual-loop configurations improve stability over the single-loop setup. The RBF-PID controller further enhances performance, achieving a settling time of 3.04 seconds, zero overshoot, and the shortest recovery time of 2.73 seconds under a 10-kg load disturbance. Although the performance gap between PI and PID remains modest, suggesting PI remains a practical solution for resource-constrained applications, the RBF-PID controller provides significant benefits in adaptability and robustness. These findings underscore the importance of adaptive pressure regulation in improving the tracking accuracy and resilience of PAM-based actuators. The choice of control strategy should therefore be guided by the specific application context, balancing control performance with computational and hardware constraints.