Leveraging U-Net and selective feature extraction for land cover classification using remote sensing imagery

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: Jan. 4, 2025

In this study, we explore an enhancement to the U-Net architecture by integrating SK-ResNeXt as encoder for Land Cover Classification (LCC) tasks using Multispectral Imaging (MSI). introduces cardinality and adaptive kernel sizes, allowing better capture multi-scale features adjust more effectively variations in spatial resolution, thereby enhancing model's ability segment complex land cover types. We evaluate approach Five-Billion-Pixels dataset, composed of 150 large-scale RGB-NIR images over 5 billion labeled pixels across 24 categories. The achieves notable improvements baseline U-Net, with gains 5.312% Overall Accuracy (OA) 8.906% mean Intersection Union (mIoU) when RGB configuration. With RG-NIR configuration, these increase 6.928% OA 6.938% mIoU, while configuration yields 5.854% 7.794% mIoU. Furthermore, not only outperforms other well-established models such DeepLabV3, DeepLabV3+, Ma-Net, SegFormer, PSPNet, particularly but also surpasses recent state-of-the-art methods. Visual tests confirmed superiority, showing that studied certain classes, lakes, rivers, industrial areas, residential vegetation, where architectures struggled achieve accurate segmentation. These results demonstrate potential capability explored handle MSI enhance LCC results.

Language: Английский

---

Association of High‐Sensitivity Troponins in Metabolic Dysfunction‐Associated Steatotic Liver Disease With All‐Cause and Cause‐Specific Mortality

Alimentary Pharmacology & Therapeutics, Journal Year: 2025, Volume and Issue: unknown

Published: April 1, 2025

Characterising the phenotypic features of individuals with metabolic dysfunction-associated steatotic liver disease (MASLD) can help identify high-risk subpopulations within this group. High-sensitivity troponin (hs-troponin) is a significant risk factor for future cardiovascular events. We studied association hs-troponin in absence all-cause and cause-specific mortality among MASLD. used National Health Nutrition Examination Survey 1999-2004 linked dataset through 2019. Cox regression models to assess between MASLD without disease. During median follow-up period 17.5 years (IQR: 15.9-19.1), higher levels T were associated progressively hazards mortality, which remained after adjustment demographic, clinical, lifestyle factors. There was 29% (hazard ratio [HR]: 1.29, 95% confidence interval [CI]: 1.16-1.44) increase 44% (HR: 1.44, CI: 1.20-1.72) every rise 1-standard deviation T. A (p trend) noted 3 I assays, similar no cancer-related mortality. Screening or at-risk group that have predominantly due disease-related population

Language: Английский

---

Discovering of ultrasound-derived fat fraction as a non-invasive and efficient identification for nonalcoholic fatty liver disease

Research Square (Research Square), Journal Year: 2025, Volume and Issue: unknown

Published: April 14, 2025

Background This study aimed to investigate ultrasound-derived fat fraction (UDFF) as a diagnostic alternative magnetic resonance imaging proton density (MRI-PDFF), the noninvasive gold standard, in metabolic dysfunction-associated steatotic liver disease (MASLD), together with comparing ability of UDFF controlled attenuation parameter (CAP). Methods The criteria for MASLD were used, we included 103 individuals 53 patients and 49 healthy controls. All participants underwent MRI measure MRI-PDFF ultrasonography CAP. receiver operating characteristic (ROC) curves determine efficacy CAP diagnosing MASLD. Finally, analyzed correlation between serological indicators consistency MRI-PDFF. Results median mean was 6%. In addition, values 246dB/m 5.44%, respectively. Overall, positively linked (R = 0.876; P < 0.001), 0.792, 0.001). For 5% above MRI-PDFF, noted have AUC 0.981. Additionally, an 0.932 Bland-Altman difference plots showed overall deviation -0.2%. A linear regression model suggested proportional error. Conclusions Ultrasound testing provides simple clinical tool quantify extent hepatic steatosis. is superior CAP, which could be

Language: Английский

---

Autonomous object tracking with vision based control using a 2DOF robotic arm

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: April 18, 2025

Abstract The tracking of moving object by implementing robot manipulator is one the challenging task for many applications such as manufacturing, agriculture, logistics, healthcare, space, military, entertainment, etc. In deployment robotic manipulators with real-time aforementioned important applications, proper sensor surveillance and ensuring stability are major challenges. purpose this study to design a precise responsive object-tracking system eliminating complexities related tedious mechanisms, rigidity, requirement multiple sensors, which commonly associated traditional systems. arms can be effectively designed track objects autonomously vision-based control. comparison different classical servoing approaches, image-based visual (IBVS) more advantageous in present article describes new approach IBVS-based control 2-degree-of-freedom (DOF) arm including identification trajectory based crucial components. To solve issues IBVS, an accurate deep learning-based detection framework employed. presented utilized detect locate real-time. Further, effective technique 2-DOF help response system. validation proposed strategy done performing simulation experimental investigations CoppeliaSim simulator arm, respectively. findings reveal that learning controller achieves good levels accuracy time while tasks. Furthermore, thorough discussion on possibility using data-driven has been explored improve robustness adaptability scheme.

Language: Английский

---