The cuticle of insect weapon structures, such as mandibles of stag beetles, plays a crucial role in their mechanical performance and evolutionary success. This study investigates the elemental composition of these cuticles, focusing on the presence of metals and other trace elements. Scanning Electron Microscopy with Energy Dispersive X-ray Spectroscopy (SEM-EDS) revealed that, in addition to the primary elements carbon (C) and oxygen (O), trace amounts of various metals were present, including nickel (Ni), which is reported here for the first time in insect cuticles. These elements are likely derived from the decaying wood consumed by larvae, emphasizing the influence of environmental factors on adult beetle morphology. The deposition of these metals is hypothesized to enhance the mechanical properties of the mandibles, contributing to their effectiveness in combat and other functions. This research advances our understanding of insect physiology and evolutionary adaptations.Biocrusts are essential components of terrestrial ecosystems. This research examines the temporal dynamic changes in microbial communities of Light Algal Crust (LAC) and Cyanobacteria Lichen Crusts (CLC) from Joshua Tree National Park (JTNP). We employed amplicon sequencing and exometabolite profiling to assess microbial community dynamics and activity during reactivation from desiccation in these crusts ranging from 3 minutes to 96 hours after wetting. The results indicated that CLC had more stable microbial community as well as metabolic activity than LAC. Conversely, LAC showed a rapid change in metabolic activity over time. These findings deepen our understanding regarding biocrust resilience by showing their capability for recovery and adjusting to shifts in their surroundings, which also give essential insights about ecological roles of biocrusts and landscaping implications for ecosystem restoration.Early detection of gastric intestinal metaplasia (GIM) is vital for preventing gastric cancer. Developing a robust real-time diagnostic system for various clinical settings, including different color modes and endoscopic ranges, remains challenging. This paper presents an innovative real-time GIM segmentation method, incorporating a switch adapter module (SAM) and a color selector (CS) for specific tasks per color mode. A high-resolution (HR) technique with half-precision compression enhances performance while maintaining fast inference speeds. Our dataset includes a comprehensive collection of esophagogastroduodenoscopy (EGD) images across diverse ranges, ensuring wide applicability. Our model surpasses existing benchmarks, achieving 94% sensitivity, 97% specificity, 93% positive predictive value, 92% negative predictive value, 98% accuracy, and a mean intersection over union (IoU) of 55%. Notably, it achieves 97.89% specificity for close-up images, demonstrating its effectiveness in clinical settings, especially at short distances.98Nut SONGVORAWIT99Nuttapon POMBUBPA100Peerapon VATEEKULAnalysis of Metal Element Deposition in the Cuticle of Beetle Weapon(Project 2021)Biocrust Reawakening: An investigation of biocrust metagenomics and exometabolomics changes after water activation(Project 2022)Automatic Gastric Intestinal Metaplasia Segmentation System from Gastroscopic Images(Project 2023)75
元のページ ../index.html#79