CFM-UNet: coupling local and global feature extraction networks for medical image segmentation.

In medical image segmentation, traditional CNN-based models excel at extracting local features but have limitations in capturing global features. Conversely, Mamba, a novel network framework, effectively captures long-range feature dependencies and excels in processing linearly arranged image inputs, albeit at the cost of overlooking fine spatial relationships and local pixel interactions. This limitation highlights the need for hybrid approaches that combine the strengths of both architectures.

Long-wave infrared computational multispectral metasurface and spectral reconstruction method.

We demonstrate a computational multispectral metasurface employing a 3×3 photonic crystal array architecture that operates across the longwave infrared spectrum (8-11.5 [Formula: see text] m). The designed structure achieves remarkable optical performance with peak transmittance reaching 75.

A large language model for multimodal identification of crop diseases and pests.

Pests and diseases significantly impact the growth and development of crops. When attempting to precisely identify disease characteristics in crop images through dialogue, existing multimodal models face numerous challenges, often leading to misinterpretation and incorrect feedback regarding disease information. This paper proposed a large language model for multimodal identification of crop diseases and pests, which can be called LLMI-CDP.

Exploring critical factors for enhancing the effectiveness of scientific and technological innovation and policy implications based on data from China.

In the context of growing global competition in science and technology, improving the effectiveness of scientific and technological (S&T) innovation is critical for strengthening China's overall S&T competitiveness. To identify the critical factors that influence the enhancement of S&T innovation effectiveness, this study employs the grounded theory and qualitative analysis techniques. Based on survey data from domestic researchers (286 responses), a critical factor model was developed consisting of five primary dimensions: "researchers' professional skills-research team collaboration-research management and organizational support-construction of a research innovation environment-research policy and incentive mechanisms".

A novel LLM time series forecasting method based on integer-decimal decomposition.

The use of traditional deep learning models for time series forecasting has demonstrated strong performance in specific domains, but their applicability remains limited due to their domain-specific nature, which restricts generalization. Inspired by advancements in natural language processing (NLP) and computer vision (CV), large language models (LLMs) have emerged as a promising method for time series forecasting. However, fundamental differences between time series data and textual data present challenges in adapting time series for LLM-based forecasting.

Multi-factor consideration sample assignment for oriented tiny object detection.

Tiny object detection in aerial image is crucial for urban planning and environmental monitoring. However, unpredictable orientation and lack of distinctive features pose challenges in sample assignment, often resulting in mismatch and inconsistency between anchors and priors. To address this, we introduce the multi-factor consideration sample assignment (MCSA) mechanism, which ensures the assignment of superior positive samples for objects with orientation.

A prediction model for soil heavy metal content based on improved tensor completion.

As socio-economic activities intensify, soil heavy metal pollution increasingly threatens both the environment and human health. This paper presents a novel method for predicting soil heavy metal content using an advanced tensor completion algorithm. The proposed method estimates heavy metal concentrations at unsampled locations by constructing a prediction model within the Coarse-to-Fine (C2F) framework, leveraging data from sampled points.

Energy consumption prediction of PEVs incorporating traffic flow information.

The problem of range anxiety caused by the discrepancy between the mileage on the dashboard and the driving mileage of pure electric vehicles (PEVs) is one of the most important reasons hindering the development of PEVs. Prediction of energy consumption can effectively reduce the driver's range anxiety and provide support for energy management strategies optimization and energy-efficient route plan. To this end, this paper analyzes the effects of velocity, environment and driving style on energy consumption from the real driving data of PEVs.

Wide-field fluorescence navigation system for efficient miniature multiphoton imaging in freely behaving animals.

Significance: Miniature multiphoton microscopy has revolutionized neuronal imaging in freely behaving animals. However, its shallow depth of field-a result of high axial resolution-combined with a limited field of view (FOV), makes it challenging for researchers to identify regions of interest in three-dimensional space across multimillimeter cranial windows, thereby reducing the system's ease of use.

Aim: We aimed to develop a multimodal imaging platform with enhanced guidance and a standardized workflow tailored for efficient imaging of freely behaving animals.

GFADE: generalized feature adaptation and discrimination enhancement for deepfake detection.

With the rapid advancement of deep generative techniques, such as generative adversarial networks (GANs), the creation of realistic fake images and videos has become increasingly accessible, raising significant security and privacy concerns. Although existing deepfake detection methods perform well within a single dataset, they often experience substantial performance degradation when applied across datasets or manipulation types. To address this challenge, we propose a novel deepfake detection framework that combines multiple loss functions and the MixStyle technique.

Diversified caching algorithm with cooperation between edge servers.

Edge computing makes up for the high latency of the central cloud network by deploying server resources in close proximity to users. The storage and other resources configured by edge servers are limited, and a reasonable cache replacement strategy is conducive to improving the cache hit ratio of edge services, thereby reducing service latency and enhancing service quality. The spatiotemporal correlation of user service request distribution brings opportunities and challenges to edge service caching.

Identification and Evolution of Core Technologies in the Chip Field Based on Patent Networks.

Currently, the global technological competition pattern is accelerating its restructuring, and chip technology, as a core technology for national strategic security and industrial competition, faces a serious bottleneck that seriously restricts the construction of China's industrial chain security and innovation ecology. A "recognition-evolution" collaborative analysis system was proposed in this study using patent data as a carrier. Firstly, a PKCN-BERT-LDA fusion module was constructed to identify the core technologies of chip design, manufacturing, and packaging testing.

Generalizable medical image enhancement using structure-preserved diffusion models.

Clinical medical images often suffer from compromised quality, which negatively impacts the diagnostic process by both clinicians and AI algorithms. While GAN-based enhancement methods have been commonly developed in recent years, delicate model training is necessary due to issues with artifacts, mode collapse, and instability. Diffusion models have shown promise in generating high-quality images superior to GANs, but challenges in training data collection and domain gaps hinder applying them for medical image enhancement.

Corrigendum to "Decomposition-integration-based prediction study on the development trend of film industry" [Heliyon Volume 9, Issue 11, November 2023, Article e21211].

[This corrects the article DOI: 10.1016/j.heliyon.

Dimensional Synthesis of 6-DOF Parallel Robot for Intra-Operative Radiation Therapy.

Background: In order to meet the kinematic requirements of large range of motion, payload, and stiffness of Intra-Operative Radiation Therapy robots, a 6-degree-of-freedom (DOF) parallel platform (Stewart-Gough mechanism) is introduced and a dimensional synthesis study is carried out.

Methods: The kinematic and static stiffness models of the 6-DOF parallel robot for Intra-Operative Radiation Therapy are derived around a virtual isocentric control point. Under the premise of ensuring the positional accuracy, the optimal dimensions of the initial rod length, the radius of the fixed base and movable platform, and the circumferential angle of the 6-DOF parallel platform are obtained by using the multi-objective optimization method combining the non-dominated sorting genetic algorithm and the global 4criterion with the working space, stiffness, and load as the optimization objectives.

Corrigendum to 'regulation of corn starch on the properties of tremella polysaccharide-egg white protein-orange juice composite gel and its application in 3D printing' [food chemistry: X 27 (2025) 102382].

[This corrects the article DOI: 10.1016/j.fochx.

The impact of autonomy-supportive organizational environments on employees' emotions and creative performance: A self-determination theory perspective.

The ongoing debate over whether positive or negative emotions foster creative performance remains a pivotal issue in understanding the interplay between emotions and creativity. Emerging research suggests that both positive and certain negative emotions, such as fear and guilt, can enhance creativity under specific conditions. Grounded in Self-Determination Theory (SDT), this study examines how autonomy-supportive organizational environments contribute to the satisfaction of employees' basic psychological needs.

Fano Resonance Mach-Zehnder Modulator Based on a Single Arm Coupled with a Photonic Crystal Nanobeam Cavity for Silicon Photonics.

Recently, Fano resonance modulators and photonic crystal nanobeam cavities (PCNCs) have attracted more and more attention due to their superior performance, such as high modulation efficiency and high extinction ratio (ER). In this paper, a silicon Fano resonance Mach-Zehnder modulator (MZM) based on a single arm coupled with a PCNC is theoretically analyzed, designed, and numerically simulated. By optimizing the coupling length, lattice constant, coupling gap, and the number of holes in the mirror/taper region, the ER of our MZM can achieve 34 dB.

Time-Interval-Guided Event Representation for Scene Understanding.

The recovery of scenes under extreme lighting conditions is pivotal for effective image analysis and feature detection. Traditional cameras face challenges with low dynamic range and limited spectral response in such scenarios. In this paper, we advocate for the adoption of event cameras to reconstruct static scenes, particularly those in low illumination.

Multi-Feature Fusion Method Based on Adaptive Dilation Convolution for Small-Object Detection.

This paper addresses the challenge of small-object detection in traffic surveillance by proposing a hybrid network architecture that combines attention mechanisms with convolutional layers. The network introduces an innovative attention mechanism into the YOLOv8 backbone, which effectively enhances the detection accuracy and robustness of small objects through fine-grained and coarse-grained attention routing on feature maps. During the feature fusion stage, we employ adaptive dilated convolution, which dynamically adjusts the dilation rate spatially based on frequency components.

Design of Low-Cost and Highly Energy-Efficient Convolutional Neural Networks Based on Deterministic Encoding.

Stochastic Computing has attracted extensive attention in the deployment of neural networks at the edge due to its low hardware cost and high fault tolerance. However, traditional stochastic computing requires a long random bit stream to achieve sufficient numerical precision. The long bit stream, in turn, increases the network inference time, hardware cost, and power consumption, which limits its application in executing tasks such as handwritten recognition, speech recognition, image processing, and image classification at the near-sensor end.

Extended Shortwave Infrared T2SL Detector Based on AlAsSb/GaSb Barrier Optimization.

Extended shortwave infrared (eSWIR) detectors operating at high temperatures are widely utilized in planetary science. A high-performance eSWIR based on pBin InAs/GaSb/AlSb type-II superlattice (T2SL) grown on a GaSb substrate is demonstrated. It achieves the optimization of the device's optoelectronic performance by adjusting the p-type doping concentration in the AlAsSb/GaSb barrier.

Improved Snake Optimization Algorithm for Global Optimization and Engineering Applications.

In engineering applications, many complex problems can be formulated as mathematical optimization challenges, and efficiently solving these problems is critical. Metaheuristic algorithms have proven highly effective in addressing a wide range of engineering issues. The Snake Optimization Algorithm (SO) is a novel metaheuristic method with widespread use.

A latent profile analysis of self-leadership associated with coping styles among general practice students.

General practice students are faced with greater stress and poorer career prospects than other medical students. Self-leadership has an important impact on their coping styles. However, understanding these relationships is complicated by the fact that six types of self-leadership strategies can be combined in various ways based on their levels and shapes.

Study on the Influence of Sn Concentration on Non-Substitutional Defect Concentration and Sn Surface Segregation in GeSn Alloys.

GeSn alloys are among the most promising materials for the fabrication of high-efficiency silicon-based light sources. However, due to the tendency of Sn to segregate to the surface during growth, it is challenging to achieve a high Sn concentration while maintaining high-quality GeSn alloys. Both theoretical and experimental studies have confirmed that non-substitutional Sn defects (VSnV) are the primary driving factors in Sn surface segregation.