In the realm of infrastructure maintenance, a novel approach to detecting concrete bridge damage has emerged. Researchers have developed an automated detection method using an enhanced version of the YOLO (You Only Look Once) algorithm, augmented with Vision Transformers. This advancement aims to improve the accuracy and efficiency of identifying structural defects in bridges, particularly those captured by drones under challenging real-world conditions. The study, published in Engineering, highlights the potential of this technology to enhance bridge inspection processes.

A new review published in Engineering explores the advancements in hand–eye coordination technology for agricultural robots, highlighting their potential to address labor shortages and enhance efficiency in complex farming tasks. The study examines various configurations of hand–eye systems, calibration methods, and control strategies, emphasizing the importance of accurate target perception and collision-free handle. While significant progress has been made, ongoing challenges such as maintaining calibration accuracy and replicating human dexterity remain. The findings suggest that continued innovation in hand–eye coordination could play a crucial role in the future development of smart agriculture.

A new study published in Engineering offers detailed insights into the potential areas for high-yield finfish mariculture. By integrating advanced physiological models with the spatiotemporal heterogeneity of marine environments, researchers have mapped suitable and high-yield areas for 27 commercial finfish species under current and future climate scenarios. The findings indicate that finfish mariculture could play a significant role in global food security as suitable areas are projected to increase by 2050.

This study published in Engineering offers new insights into the impact of alcoholic liver disease (ALD) on post-transplant outcomes in patients with hepatocellular carcinoma (HCC) and hepatitis B virus (HBV) infection. Researchers found that ALD significantly increases the risk of HBV reactivation and worsens survival outcomes following liver transplantation. Using advanced machine learning techniques, the study identified key metabolic factors associated with HBV reactivation and developed a novel risk stratification index to better predict patient outcomes. The findings highlight the importance of considering ALD in the management of liver transplant recipients with HBV-related HCC.

Researchers have developed a new hybrid earthquake early warning system called HEWFERS, which leverages advanced machine learning techniques and seismological principles to provide real-time predictions of ground shaking intensity. The system integrates a domain-informed variational autoencoder, a feed-forward neural network, and Gaussian process regression to offer both on-site and regional early warnings. Validated with a large database of ground motion records, HEWFERS demonstrates enhanced accuracy and reliability in seismic risk mitigation.

RNAi technologies have been exploited to control viruses, pests, oomycetes and fungal phytopathogens that cause disasters in host plants, including many agronomically significant crops. However, it is unclear what process mediates RNA uptake by fungi. Here, the authors utilized live-cell imaging technology combined with molecular biology experiments to demonstrate that exogenous RNA is indiscriminately absorbed by Verticillium dahliae, the notorious plant pathogenic fungus. The uptake of exogenous RNA by fungal cells is predominantly mediated through endocytosis. This study not only provides a new theoretical foundation for applying trans-kingdom RNA interference technology in crop protection but also lays the groundwork for research and applications of exogenous RNA in plant-fungi interaction systems.

The Wind Energy Research Department at the Korea Institute of Energy Research (KIER), led by Director Cheol Yoo, has developed a design platform and research-scale infrastructure to support the development of large wind turbine blades using in-house technology.

This study establishes a standardized continuous-labeling, bottom-up hydrogen deuterium exchange mass spectrometry (HDX-MS) protocol to investigate conformational dynamics in protein–small molecule interactions, using the N-terminal domain of Hsp90 (Hsp90N) as a model system.

This study focuses on lipid droplets (LDs), which are organelles responsible for storing neutral lipids and are found across various species. The excessive accumulation of these lipids in LDs is linked to several metabolic disorders. Zebrafish, with their transparent embryos, present a valuable model for studying LD biology, though previous research has been hampered by the absence of specific LD marker proteins and limitations in purification techniques.