Underwater filter-nets play a critical role in aquaculture and marine engineering, where reliable condition monitoring is essential for ensuring operational safety. Although video-based filter-net segmentation enables automated inspection and early fault detection, its performance is significantly hindered by underwater imaging challenges, including low illumination, scattering-induced visibility degradation, and pronounced spatiotemporal appearance variability. These challenges often cause conventional segmentation approaches to exhibit mask drift and error accumulation, thereby compromising stable long-term tracking. To address these challenges, we propose an enhanced SAM2-based segmentation framework incorporating two collaborative temporal-consistency mechanisms that combines mask-weakening and mask-expansion detection. The former identifies subtle structural degradation through foreground-ratio attenuation, while the latter mitigates invalid mask growth by analyzing multi-frame ratio evolution. Given the scarcity of high-quality, densely annotated underwater video datasets, we develop a comprehensively annotated underwater filter-net video segmentation dataset, UWFN. Experimental results demonstrate that our proposed approach achieves a 𝒥&ℱ score of 88.2 on the UWFN dataset, exceeding classic methods and demonstrating superior robustness in real-world underwater inspection scenarios.
Jiawei Wang, Hongwen Yu, Zini Wang et al.· International Conference on...· 0 citations
Objective Insulin resistance (IR) constitutes a central pathophysiological mechanism underlying a spectrum of chronic metabolic disorders. The Metabolic Score for Insulin Resistance (METS-IR), which integrates fasting blood glucose (FBG), triglycerides (TG), high-density lipoprotein cholesterol (HDL-C), and body mass index (BMI), serves as a simple, cost-effective surrogate marker for IR. However, its associations with type 2 diabetes mellitus (T2DM), cancer, and all-cause mortality remain to be systematically synthesized. This study aims to systematically evaluate its clinical prognostic value. Methods We systematically searched PubMed, EMBASE, and Web of Science for prospective or retrospective cohort studies evaluating the associations of baseline METS-IR with incident T2DM, cancer, or all-cause mortality. Pooled hazard ratios (HRs) and 95% confidence intervals (CIs) were calculated using random-effects models, and linear and nonlinear dose-response meta-analyses were performed. Heterogeneity, robustness, and publication bias were additionally assessed. Results A total of 23 cohort studies were included. For T2DM (6 studies), the highest METS-IR category was associated with a substantially elevated risk compared with the lowest category (HR = 4.03, 95% CI: 2.57–6.33), and per 1-standard deviation (1-SD) increment in METS-IR corresponded to a 48% higher risk (HR = 1.48, 95% CI: 1.24–1.77); dose-response analysis revealed a significant nonlinear increasing relationship (P for nonlinearity< 0.001). For cancer (5 studies), per 1-SD increment in METS-IR was associated with a 21% higher risk of cancer (HR = 1.21, 95% CI: 1.10–1.34); dose-response analysis indicated a significant nonlinear association (P for nonlinearity = 0.042). For all-cause mortality (13 studies), per 1-SD rise in METS-IR was linked to an 8% increased risk of all-cause death (HR = 1.08, 95% CI: 1.03–1.14); dose-response analysis demonstrated an overall linear increasing trend (P for linearity = 0.029), with each 20-unit increase in METS-IR associated with an approximately 2.8% higher risk of all-cause mortality (HR = 1.028, 95% CI: 1.004–1.056). Conclusions Elevated METS-IR levels are significantly associated with a nonlinear, progressive increase in the risk of incident T2DM. Its associations with cancer and all-cause mortality are weaker and more complex yet still exhibit consistent directional trends. METS-IR represents a convenient, readily applicable tool for early identification of metabolic risk using routine clinical parameters; however, its prognostic value for cancer and all-cause mortality merits further validation in large-scale, long-term follow-up studies. Systematic review registration PROSPERO, identifier CRD420261293948.
Fengwei Guo, Hui Liu, Junyi Sun et al.· Frontiers in Endocrinology· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
