Purpose: We aimed to analyze retinal neurovascular unit (RNVU) alterations and function via optical coherence tomography angiography (OCTA) full-field electroretinography (ERG) in patients with ischemic stroke (IS). Methods: OCTA was used measure RNVU changes 229 participants (101 IS 128 healthy controls). The RETeval device record electroretinograms (FERGs) 40 (14 26 Logistic regression models for were constructed. Receiver operating characteristic (ROS) curves constructed assess the...

Deploying deep neural networks (DNNs) on IoT devices for model serving is a promising solution intelligent applications with high real-time requirements and bandwidth sensitivity. To cope the prohibitive computation storage overheads of modern DNNs, great efforts have been devoted to compression technique. Most existing approaches focus minimizing size maximizing average accuracy all inference tasks. However, real-world tasks various service-level objectives (SLOs). Models compressed by...

Spectrum maps are of crucial importance for realizing efficient spectrum management in the sixth-generation (6G) wireless communication networks. However, existing map construction schemes mainly depend on spatial interpolation or just simply exploit frequency correlation and cannot accurately construct when measurement data at target not available. To overcome this challenge, we propose two accurate using different intelligent frequency-spatial reasoning methods. The among frequencies is...

Spectrum data compression with a high-rate and accurate reconstruction is of crucial importance for reducing the ultra-large transmission from edge sensors to cloud establishing high-quality spectrum maps. However, current methods ignore imbalanced edge-cloud computation resources cannot tackle outlier signals, resulting in significant challenges achieving effective compression. Therefore, we develop an efficient heterogeneous learning framework. In framework, paralleled compress normal...

The region of the RC slab–column connection is subject to complex forces and susceptible localized damage, leading progressive collapse incidents, which has raised considerable concern among engineering community. At present, research on connections exhibits several shortcomings, including a limited number specimens, incomplete consideration various factors, unclear boundaries in defining failure modes. A comprehensive punch–span ratio (a/h0) longitudinal reinforcement eigenvalues (ρƒy/ƒc)...

Automatic modulation classification (AMC) is of crucial importance for realizing wireless intelligence communications. Many deep learning based models especially convolution neural networks (CNNs) have been proposed AMC. However, the computation cost very high, which makes them inappropriate beyond fifth generation communication that stringent requirements on accuracy and computing time. In order to tackle those challenges, a novel involution enabled AMC scheme by using bottleneck structure...

Vortex-Induced Motion (VIM), which occurs as a consequence of exposure to strong current such Loop Current eddies in the Gulf Mexico, is one critical factors design mooring and riser systems for deepwater offshore structures Spars multi-column Deep Draft Floaters (DDFs). The VIM response can have significant impact on fatigue life components. In particular, Steel Catenary Risers (SCRs) suspended from floater be sensitive VIM-induced at their mudline touchdown points. Industry currently...

Precise real-time unmanned aerial vehicle (UAV) positioning is crucial for preventing unauthorized UAVs from damaging cooperative intelligent transportation systems (C-ITSs). However, UAV remains extremely challenging due to the small target size and high flexibility. Therefore, we develop a modal-oriented self-tuning fusion framework precise millimeter-wave(MMW)-vision positioning. The selects extracts cross-modal features based on modality characters, migrates Doppler of MMW radar data...

Localization problem is a significant component of the Internet Things (IoT) and interference source localization great importance in context spectrum monitoring management. However, it remains challenging to quickly but accurately locate an from distance, especially when little known about source. To handle this problem, single learning algorithm can be exploited search varying dynamics environments that make design such intractable. In our study, we employ unmanned aerial vehicle (UAV)...

Objective This study aims to identify risk factors for vascular complications during non-emergency endovascular treatment in patients with internal carotid artery occlusion (ICAO) and propose potential interventions. Method A retrospective analysis of 92 ICAO who received our center from 1 January 2018 31 June 2023, was conducted. The correlation between intraoperative studied, interaction performed. Results Our findings revealed that the use non-neurology guide wires open vessels (adjusted...

Online medical communities have revolutionized the way patients obtain medical-related information and services. Investigating what factors might influence patients’ satisfaction with doctors predicting their can help narrow down choices increase loyalty towards online communities. Considering imbalanced feature of dataset collected from Good Doctor, we integrated XGBoost SMOTE algorithm to examine these be used predict patient satisfaction. addresses issue by oversampling classification...

Interference source localization with high accuracy and time efficiency is of crucial importance for protecting spectrum resources. Due to the flexibility unmanned aerial vehicles (UAVs), exploiting UAVs locate interference has attracted intensive research interests. The off-the-shelf UAV-based schemes sources by employing UAV keep searching until it arrives at target. This obviously degrades localization. To balance localization, this paper proposes a multi-UAV-based cooperative framework...

Several recent benchmark studies have demonstrated that Computational Fluid Dynamics (CFD) is capable of capturing both nonlinear and viscous effects in offshore marine hydrodynamics predicting well certain wave- current-induced platform motion. In order to apply CFD for practical global performance analysis a complete hull-mooring-riser coupled floating system, we develop an advanced numerical wave basin combines CFD, irregular modeling, finite-element mooring modeling. Specifically, used...

In designing fixed offshore platforms located in regions of severe wave conditions, the potential resonant response hull structure due to loads must be checked. Since natural frequency vibration is typically much higher than dominant design frequency, conventional load analysis based on linear theory does not show dynamic amplification. However, it known that steep waves are nonlinear and may contain significant energy at harmonics fundamental frequency. When forcing higher-harmonic close...

In this study, a numerical wave tank was set up to simulate the free motion of Tension Leg Platform (TLP) in extreme event. For better computational efficiency, nonlinear potential flow solver is coupled with CFD software, former far-field domain and latter near-field platform motion. order benchmark against model test, five-minute time window interest selected from sea state test. The incoming irregular firstly reconstructed measured history using then applied as input simulations for two...

Based on the 3D virtual laminated element and considering beam-slab-column collaboration, this paper conducts nonlinear infinite analysis of three reinforced concrete (R. C.) space frame structural models with different seismic precautionary, respectively, after single-col umn failure collapse calculation method verification a two-bay four-floor R. C. frame. Analysis results show that: 1) increase precautionary intensity, stiffness structure when collapsing as whole increases, mode transfers...

Optimizing deep neural network (DNN) models to meet quality of service (QoS) requirements in terms accuracy and computation is crucial importance for realizing efficient on-device inference resource-constrained artificial intelligence things (AIoT). However, most existing works can hardly satisfy the aforementioned QoS since intrinsic multi-scale characteristic DNN structures has been seldom considered. In this paper, we formulate a QoS-ensured model structure optimization problem as novel...

Abstract This paper presents advanced analysis procedures for the strength and fatigue assessment of FPSO bilge keels which are developed under joint development project (JDP) between ABS Chevron. For enhanced performance FPSO/FLNG units in harsh environments, large with extended depth often considered real designs. Damage cases have been reported, but industry guidelines on structural design very limited due to nonlinearity hydrodynamic loads keels. In this paper, a first-principles...

The temperature shrinkage effect is a very important factor causing the cracking of super-long frame structures. Current design codes and recommendations for reinforced concrete (RC) structures consider influence construction process on by adopting uniform reduction coefficient neglect method. However, different processes can vary stress structure. In this paper, under cooling action, one-layer, structure with quantities indwelling times post-cast strips calculated spatial nonlinear...

In this study, we use CFD simulations to model a fully nonlinear water wave tank. Firstly, for validation purpose, regular waves with different steepness are simulated and the results compared second-order potential flow solution free surface displacement time history at fixed locations, instantaneous spatial profiles, velocity pressure fields under surface. It is shown that small steepness, solutions agree very well while large apparent differences between these two observed. The feature of...

In this study a one-way coupled fluid-structure interaction (FSI) between ocean waves and simplified offshore platform deck structure was modeled. The FSI model consists of Volume Fluid (VOF) based hydrodynamics model, structural an interface to synchronize data these two. A Computational Dynamics (CFD) analysis used capture the breaking wave impact behavior fluid on using commercially available software STAR-CCM+. 3D Finite Element (FE) developed in ABAQUS determine deflection due...