Auction-guided model diffusion for communication-efficient federated learning on non-IID data.

In 6G mobile communication systems, various AI-based network functions and applications have been standardized. Federated learning (FL) is adopted as the core learning architecture for 6G systems to avoid privacy leakage from mobile user data. However, in FL, users with non-independent and identically distributed (non-IID) datasets can deteriorate the performance of the global model because the convergence direction of the gradient for each dataset is different, thereby inducing a weight divergence problem.

Characterizing Psychiatric Disorders Through Graph Neural Networks: A Functional Connectivity Analysis of Depression and Schizophrenia.

Major depressive disorder (MDD) and schizophrenia (SZ) are among the most debilitating psychiatric disorders, characterized by widespread disruptions in large-scale brain networks. However, the commonalities and distinctions in their large-scale network distributions remain unclear. The present study aimed to leverage advanced deep learning techniques to identify these common and distinct patterns, providing insights into the shared and disorder-specific neural mechanisms underlying MDD and SZ.

Actigraphy-based step analysis for the detection of depressed mood: An explainable machine learning approach.

Introduction: Our study aims to develop an interpretable artificial intelligence (AI) model for detecting depressive symptoms using actigraphy, integrating statistically significant features into machine learning models to enhance accuracy and explainability.

Methods: We analyzed actigraphy data from 3304 participants over a one-week period, classifying them into a depressive symptom group and a non-depressive symptom group. Six machine learning models, including CatBoost (CB) and XGBoost (XGB), were trained using absolute activity indicators based on three-hour intervals, relative activity indicators (nonparametric, Cosine analysis), and daytime light intensity exposure duration.

Roles of Nanoscale Defects of Graphene in Remote Epitaxy of GaN.

Remote epitaxy through graphene enables the fabrication of freestanding membranes, facilitating the "peel-and-stack" process for semiconductor hetero-integration. While previous studies have emphasized graphene thickness, substrate bonding ionicity, and damage-free transfer of graphene for implementing remote epitaxy, the impact of nanoscale microscopic defects in graphene remains unexplored. Metal-organic chemical vapor deposition (MOCVD) of GaN requires high temperatures and a radical reaction environment, which can damage graphene.

Multimodal Alzheimer's disease recognition from image, text and audio.

Alzheimer's disease (AD) is a progressive neurodegenerative disorder that significantly affects cognitive function. One widely used diagnostic approach involves analyzing patients' verbal descriptions of pictures. While prior studies have primarily focused on speech- and text-based models, the integration of visual context is still at an early stage.

Field-programmable robotic folding sheet.

Shape transformation by folding showcases intricate geometrical change over dimension, that has long provided the embodied intelligence of autonomous systems capable of adapting to challenging environments and displaying functional versatilities. Hinge-face material assembly interfaced by shape-morphing mechanisms produced the associated means. However, the fixed hinge structure limits the accessible modes of folding configurations despite the existing capability of rectification in plant.

Disruption of hippocampal-prefrontal neural dynamics and risky decision-making in a mouse model of Alzheimer's disease.

This study investigates how amyloid pathology influences hippocampal-prefrontal neural dynamics and decision-making in Alzheimer's disease (AD) using 5XFAD mice, a well-established model system characterized by pronounced early amyloid pathology. Utilizing ecologically relevant "approach food-avoid predator" foraging tasks, we show that 5XFAD mice exhibit persistent risk-taking behaviors and reduced adaptability to changing threat conditions, indicative of impaired decision-making. Multi-regional neural recordings reveal rigid hippocampal CA1 place cell fields, decreased sharp-wave ripple (SWR) frequencies, and disrupted medial prefrontal-hippocampal connectivity, all of which correspond with deficits in behavioral flexibility during spatial risk scenarios.

Behavioral and neurochemical changes in mice induced by low-level lead exposure: Implications for ADHD and conduct disorders.

Lead (Pb) exposure in children and adolescents poses a significant public health risk due to its potential neurotoxic effects. While high-level Pb exposure is known to impair learning and cognition, the behavioral and molecular consequences of low-level Pb exposure during developmental periods remain poorly understood. This study examined behavioral and neurochemical changes in mice exposed to Pb acetate exposure via drinking water from 4 to 8 weeks of age, corresponding to the juvenile through early adult developmental stages in mice.

Noninvasive Continuous Glucose Monitoring Using Multimodal Near-Infrared, Temperature, and Pressure Signals on the Earlobe.

This study investigates a noninvasive continuous glucose monitoring (NI-CGM) system optimized for earlobe application, leveraging the site's anatomical advantages-absence of bone, muscle, and thick skin-for enhanced optical transmission. The system integrates multimodal sensing, combining near-infrared (NIR) diffuse transmission with temperature and pressure sensors. A novel Multi-Wavelength Slope Efficiency Near-Infrared Spectroscopy (MW-SE-NIRS) method is introduced, enhancing noise robustness through the slope efficiency-based parameterization of NIR signal dynamics.

Development of a digital micromirror device-based hyperspectral imaging system with dynamically adjustable measurement regions.

Hyperspectral imaging (HSI) captures both spatial and spectral information simultaneously, enabling accurate discrimination of targets that are difficult to distinguish using conventional color imaging techniques. As a result, HSI has been widely applied across various fields, including remote sensing, industrial inspection, and biomedical diagnostics. Although many HSI methods have been developed, their imaging specifications-such as spatial and spectral resolution and acquisition speed-are largely constrained by the optical components employed.

LED array-based multi-angle light scattering for aspirating smoke detection and classification.

Although smoke detectors are actively being studied to reduce false fire alarms, they still face challenging issues such as complex and elaborate alignment, high cost, large size, and poor performance. In particular, most smoke detection systems based on Mie scattering, which rely on single-scattering measurements, may not perform effectively in real-world environments where multiple scattering occurs. We present an advanced smoke detection instrument for aspirating smoke detection and classification based on multiple scattering.

Implementing Discrete Multistate Electrochemical Response to Colloidal Quantum Dots via Regulated Charge Transfer Pathways.

Facilitating or impeding charge transfer pathways enables precise control over photoluminescence (PL) intensity in quantum dots (QDs), as the transfer of charge from QDs leads to PL quenching. In this study, we achieved discrete and reversible PL intensity modulation in QDs by using electrochemical methods. By designing QD-Prussian blue (PB) composites, we leveraged PB's electroswitchable properties, where applied voltages control the oxidation state of iron ions.

High-Performance, Roll-to-Roll Fabricated Scaffold-Supported Solid Electrolyte Separator for Practical All-Solid-State Batteries.

All-solid-state batteries (ASBs) are promising candidates for next-generation energy storage systems due to their enhanced safety and potential for higher energy densities. However, achieving practical ASBs with energy densities surpassing those of state-of-the-art lithium-ion batteries (LIBs) requires the development of thin, mechanically robust solid electrolyte separators (SESs). In this study, a scalable tape casting method is employed to fabricate a thin SES with a thickness of 27 µm and a high ionic conductance of 146 mS cm.

Development of prediction models for screening depression and anxiety using smartphone and wearable-based digital phenotyping: protocol for the Smartphone and Wearable Assessment for Real-Time Screening of Depression and Anxiety (SWARTS-DA) observational study in Korea.

Introduction: Depression and anxiety are highly prevalent mental health conditions that significantly affect quality of life and cause societal burdens. However, their detection and diagnosis rates remain low owing to the limitations of the current screening methods. With rapid technological advancements and the proliferation of consumer-grade wearable devices and smartphones, their integration into digital phenotyping research has enabled the unobtrusive screening for depression and anxiety in natural settings.

DNA Hanger: Surface-Minimized Single-Molecule Immunoassay Platform.

A novel single-molecule immunoassay platform, termed DNA Hanger, is developed to address the limitations of conventional surface-based assays. By suspending biotinylated λ-phage DNA across microfabricated quartz barriers, this method enables high-specificity protein detection with minimal nonspecific binding. DNA Hanger significantly reduces background signals, achieving nonspecific binding rates as low as one protein per 236 µm of DNA.

SERS Sensor Integrated in Stretchable and Antimicrobial Wrapper for Food Quality Monitoring and Preservation.

In the wellness era, where food quality and safety are paramount, traditional food-monitoring methods, such as ribotyping and polymerase chain reaction, are often destructive and time-consuming, limiting their practicality for widespread application. Surface-enhanced Raman scattering (SERS) sensing is a promising alternative that offers real-time, nondestructive, and highly sensitive capabilities. This study proposes a nanostructured SERS sensor integrated into a stretchable and antimicrobial wrapper (NSSAW).

Thermoforming 2D films into 3D electronics for high-performance, customizable tactile sensing.

The demand for tactile sensors in robotics, virtual reality, and health care highlights the need for high performance and customizability. Despite advances in vision-based technologies, tactile sensing remains crucial for precise interaction and subtle pressure detection. In this work, we present a design and fabrication method of customizable tactile sensors based on thermoformed three-dimensional electronics.

Deep Ensemble Learning for Application Traffic Classification Using Differential Model Selection Technique.

As the Internet evolves, application traffic is becoming increasingly diverse and complex, leading network administrators to demand more accurate application traffic classification. Various deep learning-based application traffic classification methods have clearly achieved significant success, demonstrating superior classification performance compared to traditional heuristic classification approaches. However, achieving accuracy while maintaining time-efficiency and high generalization performance remains a challenge.

Wireless Organic Light-Emitting Diode Contact Lenses for On-Eye Wearable Light Sources and Their Application to Personalized Health Monitoring.

On-eye optoelectronic systems can address unmet needs across various healthcare applications, including monitoring of physiological signals related to vision or other diseases. In this context, this work introduces wearable light sources that combine ultrathin organic light-emitting diodes (OLEDs) with contact lenses. As an illustration, we demonstrate their efficacy as a robust lighting solution for electroretinography (ERG).

Low-cost large-area 100 GHz intelligent reflective surface: electrically column control of screen-printable high phase changing ratio vanadium dioxides.

In this paper, we propose for the first time 100 GHz intelligent reflective surface (IRS) using screen-printable, high phase changing ratio vanadium dioxide (VO). Sub-THz communications offer advantages such as ultra-high speed and ultra-low latency, it increases communication challenges due to path losses and non-line-of-sight (NLOS) problems. IRS is a representative solution to this NLOS problem.

Improving Sensor Adaptability and Functionality in Cartographer Simultaneous Localization and Mapping.

This paper aims to address sensor-related challenges in simultaneous localization and mapping (SLAM) systems, specifically within the open-source Google Cartographer project, which implements graph-based SLAM. The primary problem tackled is the adaptability and functionality of SLAM systems in diverse robotic applications. To solve this, we developed a novel SLAM framework that integrates five additional functionalities into the existing Google Cartographer and Robot Operating System (ROS).

Highly sensitive microdisk laser sensor for refractive index sensing via periodic meta-hole patterning.

Microdisk lasers have emerged as compact on-chip optical sensors due to their small size, simple structure, and efficient lasing capabilities. However, conventional microdisk laser sensors face challenges in enhancing interactions with external analytes, as their energy remains predominantly confined within the laser material. In this study, we present a novel microdisk laser sensor incorporating periodic meta-hole patterning, designed to enhance external interaction while maintaining the integrity of the whispering gallery mode (WGM).

Classifying Storage Temperature for Mandarin ( L.) Using Bioimpedance and Diameter Measurements with Machine Learning.

Mandarin ( L.) is consumed worldwide. Improper storage temperatures cause flavor loss and shorten shelf lives, reducing marketability.

Biocompatible Multilayered Encapsulation for Organic Light-Emitting Diodes.

Organic light-emitting diodes (OLEDs) have tremendous potential in biotechnology, but their vulnerability to oxygen and moisture presents a significant challenge in encapsulation. In this study, we developed a multilayer thin-film encapsulation consisting of dual inorganic layers and Parylene-C, offering excellent protection and biocompatibility. This encapsulation enhances the suitability of OLEDs for flexible substrates and biological applications.