Southeast University

Multimodal Transformer Integrating CT and Pathology for Postoperative Risk Stratification in Stage I Non–Small Cell Lung Cancer

A biologically interpretable multimodal artificial intelligence framework outperformed guideline-based postoperative risk stratification for stage I non–small cell lung cancer and identified patients who would benefit from adjuvant therapy.

Digital Twin Modeling and Multi-Fidelity Deep Surrogate Simulation of Corrosion Fatigue Degradation Processes for the Condition-Based Maintenance of Permanent Flexible Fall Arrest Systems

Permanent flexible fall arrest systems are safety-critical equipment whose load-bearing capacity is governed by two coupled processes acting on very different time scales: a millisecond-scale transient arrest process and a multi-year corrosion fatigue degradation process. This pa...

Study on the regulation mechanism of hydration behavior, durability and electrochemical performance of Portland cement via nanocellulose-MXene synergistic modification

Cement-based materials face a fundamental contradiction, where achieving good structural load-bearing capacity, excellent conductivity, and long-term corrosion resistance remains challenging. However, how to introduce stable conductive phases and simultaneously improve strength w...

Bridge Anomaly Diagnosis Based on a Feature-Augmented BiLSTM Autoencoder and Hotelling T2 Test

For bridges in long-term operation, only limited abnormal samples with clear labels are available, which makes it difficult to apply conventional supervised methods that require sufficient labeled data. To address this problem, this study proposes a bridge anomaly diagnosis metho...

Small-Signal Stability Enhancement of Grid-Side Converters for Permanent-Magnet Wind Turbines via Adaptive Leaky Impedance-Reshaping Compensation

Grid-following grid-side converters (GSCs) in permanent-magnet wind turbines may lose small-signal stability in weak grids because the synchronous reference frame phase-locked loop (SRF-PLL) interacts with grid impedance. For a regulated direct-current (dc) side, this paper propo...

Structural ordering, mechanical performance and reinforcement mechanism of polyvinyl alcohol-regulated calcium silicate hydrate

Calcium silicate hydrate (C-S-H) determines the mechanical properties of cementitious materials. Disordered microstructure and random pores of C-S-H significantly limit its performance. Herein, C-S-H/polyvinyl alcohol (PVA) composites were fabricated to regulate the assembly of C...

Tensile behavior of confined lap-spliced bundled-bar connections for prefabricated box retaining walls

In prefabricated box retaining walls, the corner connection region is characterized by congested reinforcement and limited construction space, making it difficult for conventional connection details to simultaneously ensure rapid assembly and reliable force transfer. To investiga...

Thermal conductivity characteristics and predictive modeling of rubber sand composite geomaterials under multiple influencing factors

Waste tire rubber has low density, high elasticity, and excellent fatigue resistance. When mixed with sand, it forms a sustainable composite geomaterial in which thermal conductivity (TC) governs thermal performance. This study systematically investigated the TC of rubber-sand co...

Multi-state switching of topological magnetic textures via enhancing ferroelectric polarization in van der Waals multiferroic heterostructure LaBr/In2Se3

Topological magnetic textures hold great promise for non-volatile memory applications, owing to their topological stability and low energy consumption. Conventional manipulations focus on the creation/annihilation of one topological state or switching between two states, however,...

Deep learning reveals ecological decoupling and rate heterogeneity in Hymenoptera wing evolution

Quantifying biomorphological evolution and integrating phenotypic data into large-scale ecological and evolutionary studies are often limited by the scalability, subjectivity, and poor interpretability of existing methods. We introduce HAIKA (Hierarchical Attention-based Interpre...

On ARX modeling with the ASYM method for errors-in-variables problems

ARX models are often used in system identification. Using high-order ARX models one can well describe arbitrary output noise. Combining high-order ARX models and model reduction techniques, several two-step identification methods are formed, such as the asymptotic method (ASYM) (...

Numerical analysis of mechanical anchors in hybrid bonded FRP strengthening system

Abstract To address the issues of unreasonable torque control and unclear failure modes in concrete anchorage systems with chemical anchors in hybrid bonded Fiber-Reinforced Polymer (FRP) strengthening system, this study focuses on the numerical simulation analysis of bolt-concre...

Bipolar Hot Carrier‐Driven Switchable Photocurrent Polarity in Plasmonic‐Functionalized Phototransistor for All‐Optical‐Configured Neuromorphic Computing

ABSTRACT Photocarrier engineering in semiconductors is critical for developing advanced optoelectronic devices. Plasmonic hot carriers enable effective modulation of the photocarrier behavior in semiconductor systems, yet their full potential is restricted by the exploitation of...

Distributed resilient event-triggered platoon control for CAVs with time-varying vehicle count under hybrid attacks

This paper investigates the distributed platoon control problem of connected automated vehicles (CAVs) subject to a time-varying vehicle count and hybrid attacks, a critical extension of conventional platoon control methods limited to fixed vehicle numbers. For the core challenge...

Natural Hazards, Critical Infrastructure Systems, and Resilience Enhancement: A Structured State-of-the-Art Review and a Synthesis-Based Rapid Assessment Framework

Abstract Critical infrastructure systems (CISs) are increasingly exposed to natural hazards whose impacts are amplified by infrastructure interdependence, urban complexity, and the growing prevalence of compound and cascading events. Existing research provides valuable insights i...

Drug Repurposing for Treating NCOR‐Related Intellectual Disability

ABSTRACT Nuclear receptor corepressors (NCOR1/2) mediate the transcriptional activity of methyl‐CpG binding protein 2 (MECP2), and genetic variants in NCOR1/2 or MECP2 associate with autism or intellectual disability. However, it is unknown whether heterozygous NCOR mutations imp...

Spmark: improving small-target text selection with smooth-pursuit gaze interaction in hybrid paper–digital environments

Precise text annotation on printed documents is essential for knowledge capture. While gaze-based interaction offers a hands-free approach, existing methods often suffer from low accuracy and ambiguous intent recognition in cluttered contexts. We propose SPMark, a hybrid gaze int...

Meningeal Lymphatic LTBR Deficiency Links Spinal Cord Injury to Secondary Brain Pathology

ABSTRACT Spinal cord injury (SCI) induces secondary brain pathology, but the mechanisms linking spinal and brain injury remain unclear. Here, we identify meningeal lymphatic dysfunction as a contributor to secondary brain pathology after SCI. In mice, meningeal lymphatic vessels...

Two-dimensional GaOOH framework materials: Ultrathin GaOOH nanosheets coordinated with organic linkers

Two-dimensional (2D) materials have attracted growing interest in applications as adsorbent due to their high surface-to-volume ratio and tunable properties. However, randomly stacked 2D structures often suffer from restricted reactant diffusion and limited access to active sites...

Balancing adversariality and naturalness in autonomous vehicle testing scenario generation via multi-objective reinforcement learning

Reinforcement learning (RL)-based adversarial scenario generation through dynamic agent–environment interactions has emerged as a promising approach for accelerating autonomous vehicle (AV) testing. However, existing RL approaches struggle to balance adversariality and naturalnes...