The space gravitational wave detection aims to detect gravitational waves in the mHz band in order to study supermassive black hole mergers,galaxy evolution and the structure of the early universe.One of its core payl...The space gravitational wave detection aims to detect gravitational waves in the mHz band in order to study supermassive black hole mergers,galaxy evolution and the structure of the early universe.One of its core payloads is a transponder-type interstellar laser interferometer,designed to measure relative displacement changes at the pico-meter level.Among its components,phasemeter is tasked with extracting the phase and frequency of the interference signal.Currently,phase-locked loop(PLL)phasemeters are commonly employed.However,the second harmonic signal generated by the mixer can restrict both the dynamic range and phase measurement accuracy of the phasemeter.This paper analyzes the interstellar laser interferometer and the impact of the second harmonic signal on the phasemeter's performance.To address these challenges,a phasemeter incorporating a second harmonic signal filter is proposed.This new design mitigates second harmonic disturbances within the phasemeter's bandwidth by dynamically adjusting the filter's cutoff frequency to track the input signal frequency,thereby suppressing the second harmonic signal in real time.Theoretical and simulation analyses demonstrate that the proposed phasemeter with a second harmonic filter significantly enhances the dynamic range.Finally,experimental results verify that the phasemeter can achieve the tracking of sudden frequency changes up to4.8 MHz.展开更多
The increasing global prevalence of mild cognitive impairment(MCI)necessitates a paradigm shift in early detection strategies.Conventional neuropsychological assessment methods,predominantly paper-and-pencil tests suc...The increasing global prevalence of mild cognitive impairment(MCI)necessitates a paradigm shift in early detection strategies.Conventional neuropsychological assessment methods,predominantly paper-and-pencil tests such as the Mini-Mental State Examination and the Montreal Cognitive Assessment,exhibit inherent limitations with respect to accessibility,administration burden,and sensitivity to subtle cognitive decline,particularly among diverse populations.This commentary critically examines a recent study that champions a novel approach:The integration of gait and handwriting kinematic parameters analyzed via machine learning for MCI screening.The present study positions itself within the broader landscape of MCI detection,with a view to comparing its advantages against established neuropsychological batteries,advanced neuroimaging(e.g.,positron emission tomography,magnetic resonance imaging),and emerging fluid biomarkers(e.g.,cerebrospinal fluid,blood-based assays).While the study demonstrates promising accuracy(74.44%area under the curve 0.74 with gait and graphic handwriting)and addresses key unmet needs in accessibility and objectivity,we highlight its cross-sectional nature,limited sample diversity,and lack of dual-task assessment as areas for future refinement.This commentary posits that kinematic biomarkers offer a distinctive,scalable,and ecologically valid approach to widespread MCI screening,thereby complementing existing methods by providing real-world functional insights.Future research should prioritize longitudinal validation,expansion to diverse cohorts,integration with multimodal data including dual-tasking,and the development of highly portable,artificial intelligence-driven solutions to achieve the democratization of early MCI detection and enable timely interventions.展开更多
Single-signal detection in orthogonal frequency-divisionmultiplexing(OFDM)systems presents a challenge due to the time-varying nature of wireless channels.Although conventional methods have limitations,particularly in...Single-signal detection in orthogonal frequency-divisionmultiplexing(OFDM)systems presents a challenge due to the time-varying nature of wireless channels.Although conventional methods have limitations,particularly inmulti-inputmultioutput orthogonal frequency divisionmultiplexing(MIMO-OFDM)systems,this paper addresses this problem by exploring advanced deep learning approaches for combined channel estimation and signal detection.Specifically,we propose two hybrid architectures that integrate a convolutional neural network(CNN)with a recurrent neural network(RNN),namely,CNN-long short-term memory(CNN-LSTM)and CNN-bidirectional-LSTM(CNNBi-LSTM),designed to enhance signal detection performance in MIMO-OFDM systems.The proposed CNN-LSTM and CNN-Bi-LSTM architectures are evaluated and compared with both traditional methods and standalone deep learning models.Training was conducted offline using a dataset generated from a 2×2 MIMO-OFDM system with a 3GPP 5G channel model.The trained models are evaluated using accuracy,loss,and computational time,and further analysis of signal detection performance is based on bit error rate,optimal cyclic prefix length,and optimal pilot subcarrier configurations under various noise conditions and channel uncertainty scenarios.The results demonstrate that the proposed CNN-based architectures,particularly the CNN-Bi-LSTM trained model,significantly reduce the need for pilot and cyclic prefix symbols while delivering superior performance,especially at SNRs.All the hybrid deep learning architectures(CNN-LSTM,CNN-Bi-LSTM)demonstrated greater robustness and adaptability under dynamic channel conditions,outperforming conventional methods and benchmark deep learning architectures.These results indicate the effectiveness of CNN-based feature extractors in learning generalized spatial patterns,positioning these hybrid models as highly efficient and reliable solutions for MIMO-OFDM signal detection in 5G and future wireless communication systems.展开更多
This paper deeply explores oversampling technology and its applications in biomedical signal detection.It first expounds on the significance of oversampling technology in biomedical signal detection,and then analyzes ...This paper deeply explores oversampling technology and its applications in biomedical signal detection.It first expounds on the significance of oversampling technology in biomedical signal detection,and then analyzes the application strategies of oversampling technology in this field.On this basis,it details the specific applications of oversampling technology in electrophysiological signal detection,biomedical imaging signal processing,and other biomedical signal detections,and verifies its effectiveness through practical case analysis,aiming to provide certain references for relevant researchers.展开更多
Atrial Fibrillation(AF)is a cardiac disorder characterized by irregular heart rhythms,typically diagnosed using Electrocardiogram(ECG)signals.In remote regions with limited healthcare personnel,automated AF detection ...Atrial Fibrillation(AF)is a cardiac disorder characterized by irregular heart rhythms,typically diagnosed using Electrocardiogram(ECG)signals.In remote regions with limited healthcare personnel,automated AF detection is extremely important.Although recent studies have explored various machine learning and deep learning approaches,challenges such as signal noise and subtle variations between AF and other cardiac rhythms continue to hinder accurate classification.In this study,we propose a novel framework that integrates robust preprocessing,comprehensive feature extraction,and an ensemble classification strategy.In the first step,ECG signals are divided into equal-sized segments using a 5-s sliding window with 50%overlap,followed by bandpass filtering between 0.5 and 45 Hz for noise removal.After preprocessing,both time and frequency-domain features are extracted,and a custom one-dimensional Convolutional Neural Network—Bidirectional Long Short-Term Memory(1D CNN-BiLSTM)architecture is introduced.Handcrafted and automated features are concatenated into a unified feature vector and classified using Support Vector Machine(SVM),Random Forest(RF),and Long Short-Term Memory(LSTM)models.A Quantum Genetic Algorithm(QGA)optimizes weighted averages of the classifier outputs for multi-class classification,distinguishing among AF,noisy,normal,and other rhythms.Evaluated on the PhysioNet 2017 Cardiology Challenge dataset,the proposed method achieved an accuracy of 94.40%and an F1-score of 92.30%,outperforming several state-of-the-art techniques.展开更多
The aim of this study was to provide insights into the current status and primary methodologies employed by marketing authorization holders(MAHs)for signal detection.These insights are intended to offer valuable refer...The aim of this study was to provide insights into the current status and primary methodologies employed by marketing authorization holders(MAHs)for signal detection.These insights are intended to offer valuable references for regulatory authorities in shaping pertinent regulatory policies.We conducted purposive sampling interviews with personnel responsible for pharmacovigilance(PV)within MAHs,in accordance with“Good Pharmacovigilance Practice(GVP)”.The interviews covered six predefined topics with open-ended discussions,including signal collection,signal detection,signal evaluation,clustered signal detection,and current challenges and issues.A total of 26 MAHs were interviewed,comprising 14 foreign-owned and 12 domestic enterprises.Foreign-owned enterprises,along with some local innovative pharmaceutical companies,utilized Oracle’s Argus and Empirica for adverse drug reaction information storage and computer-aided quantitative signal detection,respectively.The majority of domestic enterprises used the Taimei system for data storage and qualitative analysis,although a few employed other systems.Foreign-owned MAHs had comprehensive drug vigilance systems,aligning with established standards such as those of the European Union and the United States.Domestic innovative MAHs had more comprehensive drug vigilance systems compared to traditional domestic MAHs.Their signal detection approaches drew inspiration from the practices of foreign MAHs.This work synthesized findings related to the state of PV practices within these MAHs,shedding light on the challenges,achievements,and potential pathways for improvement.展开更多
A tri-signal ultrasensitive colorimetric/electrochemical detection of ovomucoid(OM)was developed.Herein,copper oxide nanoparticles(CuO NPs)were prepared,which exhibit excellent enzyme-like activity(peroxidase-like and...A tri-signal ultrasensitive colorimetric/electrochemical detection of ovomucoid(OM)was developed.Herein,copper oxide nanoparticles(CuO NPs)were prepared,which exhibit excellent enzyme-like activity(peroxidase-like and laccase-like)and electrochemical activity.CuO@3-APBA nanoparticles(CuO@3-APBA NPs)were prepared by the coordinating Cu with the amino group on 3-aminophenobenic boric acid(3-APBA)in CuO NPs.3-APBA of CuO@3-APBA can react with diol structure on sugar chain of OM under alkaline conditions.Then,a tri-signal ultrasensitive biosensing platform for OM was established based on the catalytic activity of CuO@3-APBA nanozyme.For the first signal,CuO@3-APBA can catalyze oxidation of 1,3,5-trimethylbenzene(TMB)to turn the solution from colorless to blue in the presence of H_(2)O_(2)(absorbance at 652 nm).For the second signal,CuO@3-APBA can catalyze the oxidation of substrates(2,4-dichlorophenol and 4-aminoantipyrine)and turn the solution from colorless to pink(absorbance at 510 nm).For the third signal,electrochemical oxidation peak of copper ion from Cu^(+)to Cu^(2+)of Cu O@3-APBA was recorded by differential pulse voltammetry,which was used to determine the OM.The sensing platform exhibited a wide linear range(0.0000316-100.000000 ng/mL)with a low detection limit(0.0105 pg/mL),as well as showed advantages,such as satisfactory reproducibility,good stability,and excellent selectivity.The assay has the potential applications for ultrasensitive detection of allergen in foods.展开更多
Glial fibrillary acidic protein(GFAP)is one of the discriminative biomarkers for diagnosing traumatic brain injury(TBI),and accurate determination of GFAP is clinically significant.In this study,a novel fluorescence i...Glial fibrillary acidic protein(GFAP)is one of the discriminative biomarkers for diagnosing traumatic brain injury(TBI),and accurate determination of GFAP is clinically significant.In this study,a novel fluorescence immunoassay system was designed.We encapsulated carbon dots with a high fluorescence quantum yield(QY=92.5%)inside silicon nanocapsules to serve as fluorescent markers.These markers were then integrated with the streptavidin(SA)-biotin biomagnification system and immunomagnetic separation technology for the sensitive detection of GFAP.Based on the signal cascade amplification effect of the silicon nanocapsules and SA-biotin,the fluorescence signal of the SA-biotin-modified immunofluorescence nanocapsules increased 3.6-fold compared to the carbon dot-based immunoprobe.The fluorescence immunoassay system was constructed for GFAP using SA-biotin-modified immunocapsules as the sensing probe and immunomagnetic nanoparticles as the immunorecognition probe.The fluorescence immunoassay system can specifically and ultra-sensitively quantify GFAP in blood samples,with a detection range of 10 pg/mL–10 ng/mL and detection limits of 3.2 pg/mL(serum)and 3.6 pg/mL(plasma).Moreover,the fluorescence immunoassay system exhibited prominent recoveries of 99.4%–100.4%(phosphate buffered saline),96%–102.6%(serum),and 93.2%–110.2%(plasma),with favorable specificity and excellent stabilization.The novel fluorescence immunoassay system provides a new approach to the clinical analysis of GFAP and may serve as a potential tool for screening and diagnosing TBI.展开更多
The extracellular vesicles show great potential as a noninvasive biomarker for the early detection of cancer.Hence,there is an urgent requirement to create biosensors that are time-saving,simple,and easily scalable in...The extracellular vesicles show great potential as a noninvasive biomarker for the early detection of cancer.Hence,there is an urgent requirement to create biosensors that are time-saving,simple,and easily scalable in order to accomplish rapid,sensitive,and quantitative detection of extracellular vesicles.In this study,we present a self-propelled DNA walker powered by endonuclease Nt.Bbv CI,which enables the development of a“signal on”sensing platform for the rapid and highly sensitive detection of extracellular vesicles.The DNA motor employed tracks made of streptavidin magnetic beads,which consisted of substrate strands labeled with fluorescein and motor strands locked by aptamers.The aptamer recognition of the target protein on extracellular vesicles unlocked the motor strand,initiating the DNA motor process.After replacing the optimal buffer solution containing the endonuclease Nt.BbvC I,the motor strands autonomously moved along the streptavidin magnetic beads track,continuously releasing fluorescent molecules and producing detectable fluorescence signals.Under optimal conditions,the detection range was from 2×10~4particles/mL to 2×10~9particles/mL,with a detection limit of 2.9×10~3particles/mL,demonstrating excellent selectivity.This method has demonstrated good selectivity in different tumorderived extracellular vesicles and performs well in complex biological samples.The ability to effectively analyze surface proteins of extracellular vesicles in a short period of time gives our DNA walker a tremendous potential for developing simple and cost-effective clinical diagnostic devices.展开更多
The time-varying periodic variations in Global Navigation Satellite System(GNSS)stations affect the reliable time series analysis and appropriate geophysical interpretation.In this study,we apply the singular spectrum...The time-varying periodic variations in Global Navigation Satellite System(GNSS)stations affect the reliable time series analysis and appropriate geophysical interpretation.In this study,we apply the singular spectrum analysis(SSA)method to characterize and interpret the periodic patterns of GNSS deformations in China using multiple geodetic datasets.These include 23-year observations from the Crustal Movement Observation Network of China(CMONOC),displacements inferred from the Gravity Recovery and Climate Experiment(GRACE),and loadings derived from Geophysical models(GM).The results reveal that all CMONOC time series exhibit seasonal signals characterized by amplitude and phase modulations,and the SSA method outperforms the traditional least squares fitting(LSF)method in extracting and interpreting the time-varying seasonal signals from the original time series.The decrease in the root mean square(RMS)correlates well with the annual cycle variance estimated by the SSA method,and the average reduction in noise amplitudes is nearly twice as much for SSA filtered results compared with those from the LSF method.With SSA analysis,the time-varying seasonal signals for all the selected stations can be identified in the reconstructed components corresponding to the first ten eigenvalues.Moreover,both RMS reduction and correlation analysis imply the advantages of GRACE solutions in explaining the GNSS periodic variations,and the geophysical effects can account for 71%of the GNSS annual amplitudes,and the average RMS reduction is 15%.The SSA method has proved to be useful for investigating the GNSS timevarying seasonal signals.It could be applicable as an auxiliary tool in the improvement of nonlinear variations investigations.展开更多
Large-grain REBa_(2)Cu_(3)O_(7-δ)(REBCO,RE=rare earth)bulk superconductors offer promising magnetic field trapping capabilities due to their high critical current density,making them ideal for many important applicat...Large-grain REBa_(2)Cu_(3)O_(7-δ)(REBCO,RE=rare earth)bulk superconductors offer promising magnetic field trapping capabilities due to their high critical current density,making them ideal for many important applications such as trapped field magnets.However,for such large-grain superconductor bulks,there are lots of voids and cracks forming during the process of melting preparation,and some of them can be up to hundreds of microns or even millimeters in size.Consequently,these larger size voids/cracks pose a great threat to the strength of the bulks due to the inherent brittleness of superconductor REBCO materials.In order to ensure the operational safety of related superconducting devices with bulk superconductors,it is firstly important to accurately detect these voids/cracks in them.In this paper,we proposed a method for quantitatively evaluating multiple voids/cracks in bulk superconductors through the magnetic field and displacement response signals at superconductor bulk surface.The proposed method utilizes a damage index constructed from the magnetic field signals and displacement responses to identify the number and preliminary location of multiple defects.By dividing the detection area into subdomains and combining the magnetic field signals with displacement responses within each subdomain,a particle swarm algorithm was employed to evaluate the location and size parameters of the defects.In contrast to other evaluation methods using only magnetic field or displacement response signals,the combined evaluation method using both signals can identify the number of cracks effectively.Numerical studies demonstrate that the morphology of voids and cracks reconstructed using the proposed algorithm ideally matches real defects and is applicable to cases where voids and cracks coexist.This study provides a theoretical basis for the quantitative detection of voids/cracks in bulk superconductors.展开更多
Cell competition is an evolutionarily ancient mechanism that functions to remove unfit or dangerous clonal cells in a multicellular community.A classical model is the removal of polarity-deficient clones,such as the p...Cell competition is an evolutionarily ancient mechanism that functions to remove unfit or dangerous clonal cells in a multicellular community.A classical model is the removal of polarity-deficient clones,such as the precancerous scribble(scrib)mutant clones,in Drosophila imaginal discs.The activation of Ras,Yki,or Notch signaling robustly reverses the scrib mutant clonal fate from elimination to tumorous growth.Whether these signals converge to adopt a common mechanism to overcome the elimination pressure posed by cell competition remains unclear.Using single-cell transcriptomics,we find that a critical converging point downstream of Ras,Yki,and Notch signals is the upregulation of Upd2,an IL-6 family cytokine.Overexpression of Upd2 is sufficient to rescue the scrib mutant clones from elimination.Depletion of Upd2 blocks the growth of the scrib mutant clones with active Ras,Yki,and Notch signals.Moreover,Upd2 overexpression promotes robust intestinal stem cell(ISC)proliferation,while Upd2 is intrinsically required in ISCs for the growth of the adult intestine.Together,these results identify Upd2 as a crucial cell fitness factor that sustains tissue growth but can potentiate tumorigenesis when deregulated.展开更多
Traffic sign detection is an important part of autonomous driving,and its recognition accuracy and speed are directly related to road traffic safety.Although convolutional neural networks(CNNs)have made certain breakt...Traffic sign detection is an important part of autonomous driving,and its recognition accuracy and speed are directly related to road traffic safety.Although convolutional neural networks(CNNs)have made certain breakthroughs in this field,in the face of complex scenes,such as image blur and target occlusion,the traffic sign detection continues to exhibit limited accuracy,accompanied by false positives and missed detections.To address the above problems,a traffic sign detection algorithm,You Only Look Once-based Skip Dynamic Way(YOLO-SDW)based on You Only Look Once version 8 small(YOLOv8s),is proposed.Firstly,a Skip Connection Reconstruction(SCR)module is introduced to efficiently integrate fine-grained feature information and enhance the detection accuracy of the algorithm in complex scenes.Secondly,a C2f module based on Dynamic Snake Convolution(C2f-DySnake)is proposed to dynamically adjust the receptive field information,improve the algorithm’s feature extraction ability for blurred or occluded targets,and reduce the occurrence of false detections and missed detections.Finally,the Wise Powerful IoU v2(WPIoUv2)loss function is proposed to further improve the detection accuracy of the algorithm.Experimental results show that the average precision mAP@0.5 of YOLO-SDW on the TT100K dataset is 89.2%,and mAP@0.5:0.95 is 68.5%,which is 4%and 3.3%higher than the YOLOv8s baseline,respectively.YOLO-SDW ensures real-time performance while having higher accuracy.展开更多
In this paper,we propose a novel graph signal processing convolution recurrent network(GSP CRN)for signal enhancement against high suppressive interference(HSI)in wireless communications.GSPCRN consists of the short-t...In this paper,we propose a novel graph signal processing convolution recurrent network(GSP CRN)for signal enhancement against high suppressive interference(HSI)in wireless communications.GSPCRN consists of the short-time graph signal processing(SGSP)approach and a modified convolution recurrent network.Similar to the traditional shorttime time-frequency transformation,SGSP frames the complex-valued communication signal and transforms it to the graph-domain representations,where the connection and weight flexibility of each vertex are fully taken into account.In the presence of HSI,SGSP can extract signal features from new graph-domain dimensions and empower neural networks for weak signal enhancement.Two SGSP methods,adjacency singular value decomposition and implicit graph transformation,are designed to capture relationships among the sampling points in the segmented signals.Simulation results demonstrate that our proposed GSPCRN outperforms existing classic methods in extracting weak signals from the HSI environment.When the interference-to-signal ratio exceeds 27dB,only our proposed GSPCRN can achieve the interference mitigation.展开更多
In printed circuit board(PCB)manufacturing,surface defects can significantly affect product quality.To address the performance degradation,high false detection rates,and missed detections caused by complex backgrounds...In printed circuit board(PCB)manufacturing,surface defects can significantly affect product quality.To address the performance degradation,high false detection rates,and missed detections caused by complex backgrounds in current intelligent inspection algorithms,this paper proposes CG-YOLOv8,a lightweight and improved model based on YOLOv8n for PCB surface defect detection.The proposed method optimizes the network architecture and compresses parameters to reduce model complexity while maintaining high detection accuracy,thereby enhancing the capability of identifying diverse defects under complex conditions.Specifically,a cascaded multi-receptive field(CMRF)module is adopted to replace the SPPF module in the backbone to improve feature perception,and an inverted residual mobile block(IRMB)is integrated into the C2f module to further enhance performance.Additionally,conventional convolution layers are replaced with GSConv to reduce computational cost,and a lightweight Convolutional Block Attention Module based Convolution(CBAMConv)module is introduced after Grouped Spatial Convolution(GSConv)to preserve accuracy through attention mechanisms.The detection head is also optimized by removing medium and large-scale detection layers,thereby enhancing the model’s ability to detect small-scale defects and further reducing complexity.Experimental results show that,compared to the original YOLOv8n,the proposed CG-YOLOv8 reduces parameter count by 53.9%,improves mAP@0.5 by 2.2%,and increases precision and recall by 2.0%and 1.8%,respectively.These improvements demonstrate that CG-YOLOv8 offers an efficient and lightweight solution for PCB surface defect detection.展开更多
This study proposes a lightweight rice disease detection model optimized for edge computing environments.The goal is to enhance the You Only Look Once(YOLO)v5 architecture to achieve a balance between real-time diagno...This study proposes a lightweight rice disease detection model optimized for edge computing environments.The goal is to enhance the You Only Look Once(YOLO)v5 architecture to achieve a balance between real-time diagnostic performance and computational efficiency.To this end,a total of 3234 high-resolution images(2400×1080)were collected from three major rice diseases Rice Blast,Bacterial Blight,and Brown Spot—frequently found in actual rice cultivation fields.These images served as the training dataset.The proposed YOLOv5-V2 model removes the Focus layer from the original YOLOv5s and integrates ShuffleNet V2 into the backbone,thereby resulting in both model compression and improved inference speed.Additionally,YOLOv5-P,based on PP-PicoDet,was configured as a comparative model to quantitatively evaluate performance.Experimental results demonstrated that YOLOv5-V2 achieved excellent detection performance,with an mAP 0.5 of 89.6%,mAP 0.5–0.95 of 66.7%,precision of 91.3%,and recall of 85.6%,while maintaining a lightweight model size of 6.45 MB.In contrast,YOLOv5-P exhibited a smaller model size of 4.03 MB,but showed lower performance with an mAP 0.5 of 70.3%,mAP 0.5–0.95 of 35.2%,precision of 62.3%,and recall of 74.1%.This study lays a technical foundation for the implementation of smart agriculture and real-time disease diagnosis systems by proposing a model that satisfies both accuracy and lightweight requirements.展开更多
Modern manufacturing processes have become more reliant on automation because of the accelerated transition from Industry 3.0 to Industry 4.0.Manual inspection of products on assembly lines remains inefficient,prone t...Modern manufacturing processes have become more reliant on automation because of the accelerated transition from Industry 3.0 to Industry 4.0.Manual inspection of products on assembly lines remains inefficient,prone to errors and lacks consistency,emphasizing the need for a reliable and automated inspection system.Leveraging both object detection and image segmentation approaches,this research proposes a vision-based solution for the detection of various kinds of tools in the toolkit using deep learning(DL)models.Two Intel RealSense D455f depth cameras were arranged in a top down configuration to capture both RGB and depth images of the toolkits.After applying multiple constraints and enhancing them through preprocessing and augmentation,a dataset consisting of 3300 annotated RGB-D photos was generated.Several DL models were selected through a comprehensive assessment of mean Average Precision(mAP),precision-recall equilibrium,inference latency(target≥30 FPS),and computational burden,resulting in a preference for YOLO and Region-based Convolutional Neural Networks(R-CNN)variants over ViT-based models due to the latter’s increased latency and resource requirements.YOLOV5,YOLOV8,YOLOV11,Faster R-CNN,and Mask R-CNN were trained on the annotated dataset and evaluated using key performance metrics(Recall,Accuracy,F1-score,and Precision).YOLOV11 demonstrated balanced excellence with 93.0%precision,89.9%recall,and a 90.6%F1-score in object detection,as well as 96.9%precision,95.3%recall,and a 96.5%F1-score in instance segmentation with an average inference time of 25 ms per frame(≈40 FPS),demonstrating real-time performance.Leveraging these results,a YOLOV11-based windows application was successfully deployed in a real-time assembly line environment,where it accurately processed live video streams to detect and segment tools within toolkits,demonstrating its practical effectiveness in industrial automation.The application is capable of precisely measuring socket dimensions by utilising edge detection techniques on YOLOv11 segmentation masks,in addition to detection and segmentation.This makes it possible to do specification-level quality control right on the assembly line,which improves the ability to examine things in real time.The implementation is a big step forward for intelligent manufacturing in the Industry 4.0 paradigm.It provides a scalable,efficient,and accurate way to do automated inspection and dimensional verification activities.展开更多
Objectives:Postmenopausal osteoporosis is the most common form of osteoporosis in clinical practice,affecting millions of postmenopausal women worldwide.Postmenopausal osteoporosis demands safe and effective therapies...Objectives:Postmenopausal osteoporosis is the most common form of osteoporosis in clinical practice,affecting millions of postmenopausal women worldwide.Postmenopausal osteoporosis demands safe and effective therapies.This study aimed to evaluate the potential of hederagenin(Hed)for treating osteoporosis and to elucidate its underlying mechanisms of action.Methods:The anti-osteoporotic potential of Hed was assessed by investigating its effects on ovariectomy(OVX)-induced bone loss in mice and on receptor activator of NF-kappaB ligand(RANKL)-induced osteoclast differentiation in RAW264.7 cells.Network pharmacology analysis and molecular docking were employed to identify key targets,which were subsequently validated experimentally.Results:In vitro,Hed suppressed osteoclastogenesis by inhibiting the formation of osteoclasts and F-actin rings and by down-regulating osteoclastspecific genes(Atp6v0d2 and Acp5).In vivo,Hed significantly amelioratedOVX-induced bone loss,restoring trabecular bone volume fraction(BV/TV)and trabecular number(Tb.N),while reducing trabecular separation(Tb.Sp).Network pharmacology analysis identified 142 overlapping targets linking Hed to osteoporosis,including tumor necrosis factor alpha(TNF-α),interleukin-6(IL-6),and IL-1β,with enrichment in innate immune signaling and osteoclast differentiation.Molecular docking analysis indicated strong binding affinities between Hed and targets such as TNF-α,IL-6,and IL-1β.Experimentally,Hed was found to decrease RANKL,elevate osteoprotegerin(OPG),and suppress intestinalmRNA levels of pro-inflammatory cytokines such as IL-1β,IL-6,IL-17A,and TNF-α.Conclusion:Hed exerts significant anti-osteoporotic effects inOVX-induced osteoporosis through a dualmechanism involving the suppression of both osteoclastogenesis and innate immune signaling pathways.These findings highlighted Hed’s novel role in modulating immune-bone crosstalk,offering a promising strategy for treating osteolytic diseases without estrogenic side effects.展开更多
Synthetic speech detection is an essential task in the field of voice security,aimed at identifying deceptive voice attacks generated by text-to-speech(TTS)systems or voice conversion(VC)systems.In this paper,we propo...Synthetic speech detection is an essential task in the field of voice security,aimed at identifying deceptive voice attacks generated by text-to-speech(TTS)systems or voice conversion(VC)systems.In this paper,we propose a synthetic speech detection model called TFTransformer,which integrates both local and global features to enhance detection capabilities by effectively modeling local and global dependencies.Structurally,the model is divided into two main components:a front-end and a back-end.The front-end of the model uses a combination of SincLayer and two-dimensional(2D)convolution to extract high-level feature maps(HFM)containing local dependency of the input speech signals.The back-end uses time-frequency Transformer module to process these feature maps and further capture global dependency.Furthermore,we propose TFTransformer-SE,which incorporates a channel attention mechanism within the 2D convolutional blocks.This enhancement aims to more effectively capture local dependencies,thereby improving the model’s performance.The experiments were conducted on the ASVspoof 2021 LA dataset,and the results showed that the model achieved an equal error rate(EER)of 3.37%without data augmentation.Additionally,we evaluated the model using the ASVspoof 2019 LA dataset,achieving an EER of 0.84%,also without data augmentation.This demonstrates that combining local and global dependencies in the time-frequency domain can significantly improve detection accuracy.展开更多
We theoretically investigate the phase sensitivity of a truncated SU(1,1)interferometer fed with a two-mode coherent state and employing double-port homodyne detection.On the one hand,we analytically demonstrate that ...We theoretically investigate the phase sensitivity of a truncated SU(1,1)interferometer fed with a two-mode coherent state and employing double-port homodyne detection.On the one hand,we analytically demonstrate that the two-mode coherent state provides better phase sensitivity than the single-mode coherent state.In addition,we show that the doubleport homodyne detection is a quasi-optimal measurement.For a bright coherent-state input,the sensitivity of this scheme saturates the phase-sensitivity bound determined by the quantum Fisher information.On the other hand,we quantitatively illustrate the advantage of double-port homodyne detection over the single-port scheme under ideal conditions and in the presence of photon loss,respectively.Furthermore,our analysis indicates that the scheme we propose is robust against photon loss.展开更多
基金the National Key Research&Development Program of China(Grant No.2022YFC2203901)the State Key Laboratory of Spatial Datum(Grant No.SKLSD2025-KF-03)+1 种基金Fundamental Research Funds for the Central UniversitiesSun Yat-sen University for the support。
文摘The space gravitational wave detection aims to detect gravitational waves in the mHz band in order to study supermassive black hole mergers,galaxy evolution and the structure of the early universe.One of its core payloads is a transponder-type interstellar laser interferometer,designed to measure relative displacement changes at the pico-meter level.Among its components,phasemeter is tasked with extracting the phase and frequency of the interference signal.Currently,phase-locked loop(PLL)phasemeters are commonly employed.However,the second harmonic signal generated by the mixer can restrict both the dynamic range and phase measurement accuracy of the phasemeter.This paper analyzes the interstellar laser interferometer and the impact of the second harmonic signal on the phasemeter's performance.To address these challenges,a phasemeter incorporating a second harmonic signal filter is proposed.This new design mitigates second harmonic disturbances within the phasemeter's bandwidth by dynamically adjusting the filter's cutoff frequency to track the input signal frequency,thereby suppressing the second harmonic signal in real time.Theoretical and simulation analyses demonstrate that the proposed phasemeter with a second harmonic filter significantly enhances the dynamic range.Finally,experimental results verify that the phasemeter can achieve the tracking of sudden frequency changes up to4.8 MHz.
文摘The increasing global prevalence of mild cognitive impairment(MCI)necessitates a paradigm shift in early detection strategies.Conventional neuropsychological assessment methods,predominantly paper-and-pencil tests such as the Mini-Mental State Examination and the Montreal Cognitive Assessment,exhibit inherent limitations with respect to accessibility,administration burden,and sensitivity to subtle cognitive decline,particularly among diverse populations.This commentary critically examines a recent study that champions a novel approach:The integration of gait and handwriting kinematic parameters analyzed via machine learning for MCI screening.The present study positions itself within the broader landscape of MCI detection,with a view to comparing its advantages against established neuropsychological batteries,advanced neuroimaging(e.g.,positron emission tomography,magnetic resonance imaging),and emerging fluid biomarkers(e.g.,cerebrospinal fluid,blood-based assays).While the study demonstrates promising accuracy(74.44%area under the curve 0.74 with gait and graphic handwriting)and addresses key unmet needs in accessibility and objectivity,we highlight its cross-sectional nature,limited sample diversity,and lack of dual-task assessment as areas for future refinement.This commentary posits that kinematic biomarkers offer a distinctive,scalable,and ecologically valid approach to widespread MCI screening,thereby complementing existing methods by providing real-world functional insights.Future research should prioritize longitudinal validation,expansion to diverse cohorts,integration with multimodal data including dual-tasking,and the development of highly portable,artificial intelligence-driven solutions to achieve the democratization of early MCI detection and enable timely interventions.
基金supported by the IITP(Institute of Information&Communications Technology Planning&Evaluation)-ICAN(ICT Challenge and Advanced Network of HRD)grant funded by the Korea government(Ministry of Science and ICT)(IITP-2025-RS-2022-00156299).
文摘Single-signal detection in orthogonal frequency-divisionmultiplexing(OFDM)systems presents a challenge due to the time-varying nature of wireless channels.Although conventional methods have limitations,particularly inmulti-inputmultioutput orthogonal frequency divisionmultiplexing(MIMO-OFDM)systems,this paper addresses this problem by exploring advanced deep learning approaches for combined channel estimation and signal detection.Specifically,we propose two hybrid architectures that integrate a convolutional neural network(CNN)with a recurrent neural network(RNN),namely,CNN-long short-term memory(CNN-LSTM)and CNN-bidirectional-LSTM(CNNBi-LSTM),designed to enhance signal detection performance in MIMO-OFDM systems.The proposed CNN-LSTM and CNN-Bi-LSTM architectures are evaluated and compared with both traditional methods and standalone deep learning models.Training was conducted offline using a dataset generated from a 2×2 MIMO-OFDM system with a 3GPP 5G channel model.The trained models are evaluated using accuracy,loss,and computational time,and further analysis of signal detection performance is based on bit error rate,optimal cyclic prefix length,and optimal pilot subcarrier configurations under various noise conditions and channel uncertainty scenarios.The results demonstrate that the proposed CNN-based architectures,particularly the CNN-Bi-LSTM trained model,significantly reduce the need for pilot and cyclic prefix symbols while delivering superior performance,especially at SNRs.All the hybrid deep learning architectures(CNN-LSTM,CNN-Bi-LSTM)demonstrated greater robustness and adaptability under dynamic channel conditions,outperforming conventional methods and benchmark deep learning architectures.These results indicate the effectiveness of CNN-based feature extractors in learning generalized spatial patterns,positioning these hybrid models as highly efficient and reliable solutions for MIMO-OFDM signal detection in 5G and future wireless communication systems.
文摘This paper deeply explores oversampling technology and its applications in biomedical signal detection.It first expounds on the significance of oversampling technology in biomedical signal detection,and then analyzes the application strategies of oversampling technology in this field.On this basis,it details the specific applications of oversampling technology in electrophysiological signal detection,biomedical imaging signal processing,and other biomedical signal detections,and verifies its effectiveness through practical case analysis,aiming to provide certain references for relevant researchers.
基金supported and funded by the Deanship of Scientific Research at Imam Mohammad Ibn Saud Islamic University(IMSIU)(grant number IMSIU-DDRSP2501)。
文摘Atrial Fibrillation(AF)is a cardiac disorder characterized by irregular heart rhythms,typically diagnosed using Electrocardiogram(ECG)signals.In remote regions with limited healthcare personnel,automated AF detection is extremely important.Although recent studies have explored various machine learning and deep learning approaches,challenges such as signal noise and subtle variations between AF and other cardiac rhythms continue to hinder accurate classification.In this study,we propose a novel framework that integrates robust preprocessing,comprehensive feature extraction,and an ensemble classification strategy.In the first step,ECG signals are divided into equal-sized segments using a 5-s sliding window with 50%overlap,followed by bandpass filtering between 0.5 and 45 Hz for noise removal.After preprocessing,both time and frequency-domain features are extracted,and a custom one-dimensional Convolutional Neural Network—Bidirectional Long Short-Term Memory(1D CNN-BiLSTM)architecture is introduced.Handcrafted and automated features are concatenated into a unified feature vector and classified using Support Vector Machine(SVM),Random Forest(RF),and Long Short-Term Memory(LSTM)models.A Quantum Genetic Algorithm(QGA)optimizes weighted averages of the classifier outputs for multi-class classification,distinguishing among AF,noisy,normal,and other rhythms.Evaluated on the PhysioNet 2017 Cardiology Challenge dataset,the proposed method achieved an accuracy of 94.40%and an F1-score of 92.30%,outperforming several state-of-the-art techniques.
基金China National Key Research and Development Program(Grant No.2020YFC2009000 and 2020YFC2009001)National High Level Hospital Clinical Research Funding(Grant No.BJ-2023-200)the project funding from the Drug Evaluation Center of the National Medical Products Administration(Grant No.CDRW20214001).
文摘The aim of this study was to provide insights into the current status and primary methodologies employed by marketing authorization holders(MAHs)for signal detection.These insights are intended to offer valuable references for regulatory authorities in shaping pertinent regulatory policies.We conducted purposive sampling interviews with personnel responsible for pharmacovigilance(PV)within MAHs,in accordance with“Good Pharmacovigilance Practice(GVP)”.The interviews covered six predefined topics with open-ended discussions,including signal collection,signal detection,signal evaluation,clustered signal detection,and current challenges and issues.A total of 26 MAHs were interviewed,comprising 14 foreign-owned and 12 domestic enterprises.Foreign-owned enterprises,along with some local innovative pharmaceutical companies,utilized Oracle’s Argus and Empirica for adverse drug reaction information storage and computer-aided quantitative signal detection,respectively.The majority of domestic enterprises used the Taimei system for data storage and qualitative analysis,although a few employed other systems.Foreign-owned MAHs had comprehensive drug vigilance systems,aligning with established standards such as those of the European Union and the United States.Domestic innovative MAHs had more comprehensive drug vigilance systems compared to traditional domestic MAHs.Their signal detection approaches drew inspiration from the practices of foreign MAHs.This work synthesized findings related to the state of PV practices within these MAHs,shedding light on the challenges,achievements,and potential pathways for improvement.
基金financially supported by Major Science and Technology Project of Yunnan Province(202302AE090022)the National Natural Science Foundation of China(21764005,32160236)+4 种基金National Key Research and Development Program of China(2022YFC2601604)Spring City Plan:the High-level Talent Promotion and Training Project of Kunming(2022SCP001)the Second Phase of“Double-First Class”Program Construction of Yunnan UniversityProgram for Donglu Scholars of Yunnan Universitythe Research Innovation Fund for Graduate Students of School of Chemical Science and Technology,Yunnan University。
文摘A tri-signal ultrasensitive colorimetric/electrochemical detection of ovomucoid(OM)was developed.Herein,copper oxide nanoparticles(CuO NPs)were prepared,which exhibit excellent enzyme-like activity(peroxidase-like and laccase-like)and electrochemical activity.CuO@3-APBA nanoparticles(CuO@3-APBA NPs)were prepared by the coordinating Cu with the amino group on 3-aminophenobenic boric acid(3-APBA)in CuO NPs.3-APBA of CuO@3-APBA can react with diol structure on sugar chain of OM under alkaline conditions.Then,a tri-signal ultrasensitive biosensing platform for OM was established based on the catalytic activity of CuO@3-APBA nanozyme.For the first signal,CuO@3-APBA can catalyze oxidation of 1,3,5-trimethylbenzene(TMB)to turn the solution from colorless to blue in the presence of H_(2)O_(2)(absorbance at 652 nm).For the second signal,CuO@3-APBA can catalyze the oxidation of substrates(2,4-dichlorophenol and 4-aminoantipyrine)and turn the solution from colorless to pink(absorbance at 510 nm).For the third signal,electrochemical oxidation peak of copper ion from Cu^(+)to Cu^(2+)of Cu O@3-APBA was recorded by differential pulse voltammetry,which was used to determine the OM.The sensing platform exhibited a wide linear range(0.0000316-100.000000 ng/mL)with a low detection limit(0.0105 pg/mL),as well as showed advantages,such as satisfactory reproducibility,good stability,and excellent selectivity.The assay has the potential applications for ultrasensitive detection of allergen in foods.
基金supported by the AMS Funding Project(No.ZZB2023C7010).
文摘Glial fibrillary acidic protein(GFAP)is one of the discriminative biomarkers for diagnosing traumatic brain injury(TBI),and accurate determination of GFAP is clinically significant.In this study,a novel fluorescence immunoassay system was designed.We encapsulated carbon dots with a high fluorescence quantum yield(QY=92.5%)inside silicon nanocapsules to serve as fluorescent markers.These markers were then integrated with the streptavidin(SA)-biotin biomagnification system and immunomagnetic separation technology for the sensitive detection of GFAP.Based on the signal cascade amplification effect of the silicon nanocapsules and SA-biotin,the fluorescence signal of the SA-biotin-modified immunofluorescence nanocapsules increased 3.6-fold compared to the carbon dot-based immunoprobe.The fluorescence immunoassay system was constructed for GFAP using SA-biotin-modified immunocapsules as the sensing probe and immunomagnetic nanoparticles as the immunorecognition probe.The fluorescence immunoassay system can specifically and ultra-sensitively quantify GFAP in blood samples,with a detection range of 10 pg/mL–10 ng/mL and detection limits of 3.2 pg/mL(serum)and 3.6 pg/mL(plasma).Moreover,the fluorescence immunoassay system exhibited prominent recoveries of 99.4%–100.4%(phosphate buffered saline),96%–102.6%(serum),and 93.2%–110.2%(plasma),with favorable specificity and excellent stabilization.The novel fluorescence immunoassay system provides a new approach to the clinical analysis of GFAP and may serve as a potential tool for screening and diagnosing TBI.
基金supported financially by the National Natural Science Foundation of China(NSFC,Nos.62071119 and 62075098)the National Key Research and Development Program of China(Nos.2017YFA0205301 and 2018YFC1602905)。
文摘The extracellular vesicles show great potential as a noninvasive biomarker for the early detection of cancer.Hence,there is an urgent requirement to create biosensors that are time-saving,simple,and easily scalable in order to accomplish rapid,sensitive,and quantitative detection of extracellular vesicles.In this study,we present a self-propelled DNA walker powered by endonuclease Nt.Bbv CI,which enables the development of a“signal on”sensing platform for the rapid and highly sensitive detection of extracellular vesicles.The DNA motor employed tracks made of streptavidin magnetic beads,which consisted of substrate strands labeled with fluorescein and motor strands locked by aptamers.The aptamer recognition of the target protein on extracellular vesicles unlocked the motor strand,initiating the DNA motor process.After replacing the optimal buffer solution containing the endonuclease Nt.BbvC I,the motor strands autonomously moved along the streptavidin magnetic beads track,continuously releasing fluorescent molecules and producing detectable fluorescence signals.Under optimal conditions,the detection range was from 2×10~4particles/mL to 2×10~9particles/mL,with a detection limit of 2.9×10~3particles/mL,demonstrating excellent selectivity.This method has demonstrated good selectivity in different tumorderived extracellular vesicles and performs well in complex biological samples.The ability to effectively analyze surface proteins of extracellular vesicles in a short period of time gives our DNA walker a tremendous potential for developing simple and cost-effective clinical diagnostic devices.
基金supported by the National Natural Science Foundation of China(NO.42104028,42174030 and 42004017)the Open Fund of Hubei Luojia Laboratory(No.220100048 and 230100021)the Scientific Research Project of Hubei Provincial Department of Education,and Research Foundation of the Department of Natural Resources of Hunan Province(No.20230104CH)。
文摘The time-varying periodic variations in Global Navigation Satellite System(GNSS)stations affect the reliable time series analysis and appropriate geophysical interpretation.In this study,we apply the singular spectrum analysis(SSA)method to characterize and interpret the periodic patterns of GNSS deformations in China using multiple geodetic datasets.These include 23-year observations from the Crustal Movement Observation Network of China(CMONOC),displacements inferred from the Gravity Recovery and Climate Experiment(GRACE),and loadings derived from Geophysical models(GM).The results reveal that all CMONOC time series exhibit seasonal signals characterized by amplitude and phase modulations,and the SSA method outperforms the traditional least squares fitting(LSF)method in extracting and interpreting the time-varying seasonal signals from the original time series.The decrease in the root mean square(RMS)correlates well with the annual cycle variance estimated by the SSA method,and the average reduction in noise amplitudes is nearly twice as much for SSA filtered results compared with those from the LSF method.With SSA analysis,the time-varying seasonal signals for all the selected stations can be identified in the reconstructed components corresponding to the first ten eigenvalues.Moreover,both RMS reduction and correlation analysis imply the advantages of GRACE solutions in explaining the GNSS periodic variations,and the geophysical effects can account for 71%of the GNSS annual amplitudes,and the average RMS reduction is 15%.The SSA method has proved to be useful for investigating the GNSS timevarying seasonal signals.It could be applicable as an auxiliary tool in the improvement of nonlinear variations investigations.
基金supported by the National Natural Science Foundation of China(Grant Nos.12232005 and 12072101).
文摘Large-grain REBa_(2)Cu_(3)O_(7-δ)(REBCO,RE=rare earth)bulk superconductors offer promising magnetic field trapping capabilities due to their high critical current density,making them ideal for many important applications such as trapped field magnets.However,for such large-grain superconductor bulks,there are lots of voids and cracks forming during the process of melting preparation,and some of them can be up to hundreds of microns or even millimeters in size.Consequently,these larger size voids/cracks pose a great threat to the strength of the bulks due to the inherent brittleness of superconductor REBCO materials.In order to ensure the operational safety of related superconducting devices with bulk superconductors,it is firstly important to accurately detect these voids/cracks in them.In this paper,we proposed a method for quantitatively evaluating multiple voids/cracks in bulk superconductors through the magnetic field and displacement response signals at superconductor bulk surface.The proposed method utilizes a damage index constructed from the magnetic field signals and displacement responses to identify the number and preliminary location of multiple defects.By dividing the detection area into subdomains and combining the magnetic field signals with displacement responses within each subdomain,a particle swarm algorithm was employed to evaluate the location and size parameters of the defects.In contrast to other evaluation methods using only magnetic field or displacement response signals,the combined evaluation method using both signals can identify the number of cracks effectively.Numerical studies demonstrate that the morphology of voids and cracks reconstructed using the proposed algorithm ideally matches real defects and is applicable to cases where voids and cracks coexist.This study provides a theoretical basis for the quantitative detection of voids/cracks in bulk superconductors.
基金supported by grants to Yan Yan from the Research Grants Council of the Hong Kong Special Administrative Region(GRF16103620,GRF16104324,T13-602/21N)from Shenzhen Science and Technology Innovation Commission(JCYJ20200109140201722)+1 种基金to Toyotaka Ishibashi from the National Natural Science Foundation of China(32170548)to Zongzhao Zhai from the National Natural Science Foundation of China(32170509 and 31871469).
文摘Cell competition is an evolutionarily ancient mechanism that functions to remove unfit or dangerous clonal cells in a multicellular community.A classical model is the removal of polarity-deficient clones,such as the precancerous scribble(scrib)mutant clones,in Drosophila imaginal discs.The activation of Ras,Yki,or Notch signaling robustly reverses the scrib mutant clonal fate from elimination to tumorous growth.Whether these signals converge to adopt a common mechanism to overcome the elimination pressure posed by cell competition remains unclear.Using single-cell transcriptomics,we find that a critical converging point downstream of Ras,Yki,and Notch signals is the upregulation of Upd2,an IL-6 family cytokine.Overexpression of Upd2 is sufficient to rescue the scrib mutant clones from elimination.Depletion of Upd2 blocks the growth of the scrib mutant clones with active Ras,Yki,and Notch signals.Moreover,Upd2 overexpression promotes robust intestinal stem cell(ISC)proliferation,while Upd2 is intrinsically required in ISCs for the growth of the adult intestine.Together,these results identify Upd2 as a crucial cell fitness factor that sustains tissue growth but can potentiate tumorigenesis when deregulated.
基金funded by Key research and development Program of Henan Province(No.251111211200)National Natural Science Foundation of China(Grant No.U2004163).
文摘Traffic sign detection is an important part of autonomous driving,and its recognition accuracy and speed are directly related to road traffic safety.Although convolutional neural networks(CNNs)have made certain breakthroughs in this field,in the face of complex scenes,such as image blur and target occlusion,the traffic sign detection continues to exhibit limited accuracy,accompanied by false positives and missed detections.To address the above problems,a traffic sign detection algorithm,You Only Look Once-based Skip Dynamic Way(YOLO-SDW)based on You Only Look Once version 8 small(YOLOv8s),is proposed.Firstly,a Skip Connection Reconstruction(SCR)module is introduced to efficiently integrate fine-grained feature information and enhance the detection accuracy of the algorithm in complex scenes.Secondly,a C2f module based on Dynamic Snake Convolution(C2f-DySnake)is proposed to dynamically adjust the receptive field information,improve the algorithm’s feature extraction ability for blurred or occluded targets,and reduce the occurrence of false detections and missed detections.Finally,the Wise Powerful IoU v2(WPIoUv2)loss function is proposed to further improve the detection accuracy of the algorithm.Experimental results show that the average precision mAP@0.5 of YOLO-SDW on the TT100K dataset is 89.2%,and mAP@0.5:0.95 is 68.5%,which is 4%and 3.3%higher than the YOLOv8s baseline,respectively.YOLO-SDW ensures real-time performance while having higher accuracy.
基金supported by he National Social Science Found of China(2022-SKJJ-B-112).
文摘In this paper,we propose a novel graph signal processing convolution recurrent network(GSP CRN)for signal enhancement against high suppressive interference(HSI)in wireless communications.GSPCRN consists of the short-time graph signal processing(SGSP)approach and a modified convolution recurrent network.Similar to the traditional shorttime time-frequency transformation,SGSP frames the complex-valued communication signal and transforms it to the graph-domain representations,where the connection and weight flexibility of each vertex are fully taken into account.In the presence of HSI,SGSP can extract signal features from new graph-domain dimensions and empower neural networks for weak signal enhancement.Two SGSP methods,adjacency singular value decomposition and implicit graph transformation,are designed to capture relationships among the sampling points in the segmented signals.Simulation results demonstrate that our proposed GSPCRN outperforms existing classic methods in extracting weak signals from the HSI environment.When the interference-to-signal ratio exceeds 27dB,only our proposed GSPCRN can achieve the interference mitigation.
基金funded by the Joint Funds of the National Natural Science Foundation of China(U2341223)the Beijing Municipal Natural Science Foundation(No.4232067).
文摘In printed circuit board(PCB)manufacturing,surface defects can significantly affect product quality.To address the performance degradation,high false detection rates,and missed detections caused by complex backgrounds in current intelligent inspection algorithms,this paper proposes CG-YOLOv8,a lightweight and improved model based on YOLOv8n for PCB surface defect detection.The proposed method optimizes the network architecture and compresses parameters to reduce model complexity while maintaining high detection accuracy,thereby enhancing the capability of identifying diverse defects under complex conditions.Specifically,a cascaded multi-receptive field(CMRF)module is adopted to replace the SPPF module in the backbone to improve feature perception,and an inverted residual mobile block(IRMB)is integrated into the C2f module to further enhance performance.Additionally,conventional convolution layers are replaced with GSConv to reduce computational cost,and a lightweight Convolutional Block Attention Module based Convolution(CBAMConv)module is introduced after Grouped Spatial Convolution(GSConv)to preserve accuracy through attention mechanisms.The detection head is also optimized by removing medium and large-scale detection layers,thereby enhancing the model’s ability to detect small-scale defects and further reducing complexity.Experimental results show that,compared to the original YOLOv8n,the proposed CG-YOLOv8 reduces parameter count by 53.9%,improves mAP@0.5 by 2.2%,and increases precision and recall by 2.0%and 1.8%,respectively.These improvements demonstrate that CG-YOLOv8 offers an efficient and lightweight solution for PCB surface defect detection.
文摘This study proposes a lightweight rice disease detection model optimized for edge computing environments.The goal is to enhance the You Only Look Once(YOLO)v5 architecture to achieve a balance between real-time diagnostic performance and computational efficiency.To this end,a total of 3234 high-resolution images(2400×1080)were collected from three major rice diseases Rice Blast,Bacterial Blight,and Brown Spot—frequently found in actual rice cultivation fields.These images served as the training dataset.The proposed YOLOv5-V2 model removes the Focus layer from the original YOLOv5s and integrates ShuffleNet V2 into the backbone,thereby resulting in both model compression and improved inference speed.Additionally,YOLOv5-P,based on PP-PicoDet,was configured as a comparative model to quantitatively evaluate performance.Experimental results demonstrated that YOLOv5-V2 achieved excellent detection performance,with an mAP 0.5 of 89.6%,mAP 0.5–0.95 of 66.7%,precision of 91.3%,and recall of 85.6%,while maintaining a lightweight model size of 6.45 MB.In contrast,YOLOv5-P exhibited a smaller model size of 4.03 MB,but showed lower performance with an mAP 0.5 of 70.3%,mAP 0.5–0.95 of 35.2%,precision of 62.3%,and recall of 74.1%.This study lays a technical foundation for the implementation of smart agriculture and real-time disease diagnosis systems by proposing a model that satisfies both accuracy and lightweight requirements.
文摘Modern manufacturing processes have become more reliant on automation because of the accelerated transition from Industry 3.0 to Industry 4.0.Manual inspection of products on assembly lines remains inefficient,prone to errors and lacks consistency,emphasizing the need for a reliable and automated inspection system.Leveraging both object detection and image segmentation approaches,this research proposes a vision-based solution for the detection of various kinds of tools in the toolkit using deep learning(DL)models.Two Intel RealSense D455f depth cameras were arranged in a top down configuration to capture both RGB and depth images of the toolkits.After applying multiple constraints and enhancing them through preprocessing and augmentation,a dataset consisting of 3300 annotated RGB-D photos was generated.Several DL models were selected through a comprehensive assessment of mean Average Precision(mAP),precision-recall equilibrium,inference latency(target≥30 FPS),and computational burden,resulting in a preference for YOLO and Region-based Convolutional Neural Networks(R-CNN)variants over ViT-based models due to the latter’s increased latency and resource requirements.YOLOV5,YOLOV8,YOLOV11,Faster R-CNN,and Mask R-CNN were trained on the annotated dataset and evaluated using key performance metrics(Recall,Accuracy,F1-score,and Precision).YOLOV11 demonstrated balanced excellence with 93.0%precision,89.9%recall,and a 90.6%F1-score in object detection,as well as 96.9%precision,95.3%recall,and a 96.5%F1-score in instance segmentation with an average inference time of 25 ms per frame(≈40 FPS),demonstrating real-time performance.Leveraging these results,a YOLOV11-based windows application was successfully deployed in a real-time assembly line environment,where it accurately processed live video streams to detect and segment tools within toolkits,demonstrating its practical effectiveness in industrial automation.The application is capable of precisely measuring socket dimensions by utilising edge detection techniques on YOLOv11 segmentation masks,in addition to detection and segmentation.This makes it possible to do specification-level quality control right on the assembly line,which improves the ability to examine things in real time.The implementation is a big step forward for intelligent manufacturing in the Industry 4.0 paradigm.It provides a scalable,efficient,and accurate way to do automated inspection and dimensional verification activities.
基金supported by the Scientific Research Project of Anhui ProvincialHealth Commission(Grant No.AHWJ2021b063)National Natural Scientific Foundation of China(Grant No.82160048)+1 种基金Natural Science Foundation Project of Anhui Province(Grant No.2308085MH265)Major Scientific Research Project of Anhui Provincial Department of Education(Grant No.2024AH040205).
文摘Objectives:Postmenopausal osteoporosis is the most common form of osteoporosis in clinical practice,affecting millions of postmenopausal women worldwide.Postmenopausal osteoporosis demands safe and effective therapies.This study aimed to evaluate the potential of hederagenin(Hed)for treating osteoporosis and to elucidate its underlying mechanisms of action.Methods:The anti-osteoporotic potential of Hed was assessed by investigating its effects on ovariectomy(OVX)-induced bone loss in mice and on receptor activator of NF-kappaB ligand(RANKL)-induced osteoclast differentiation in RAW264.7 cells.Network pharmacology analysis and molecular docking were employed to identify key targets,which were subsequently validated experimentally.Results:In vitro,Hed suppressed osteoclastogenesis by inhibiting the formation of osteoclasts and F-actin rings and by down-regulating osteoclastspecific genes(Atp6v0d2 and Acp5).In vivo,Hed significantly amelioratedOVX-induced bone loss,restoring trabecular bone volume fraction(BV/TV)and trabecular number(Tb.N),while reducing trabecular separation(Tb.Sp).Network pharmacology analysis identified 142 overlapping targets linking Hed to osteoporosis,including tumor necrosis factor alpha(TNF-α),interleukin-6(IL-6),and IL-1β,with enrichment in innate immune signaling and osteoclast differentiation.Molecular docking analysis indicated strong binding affinities between Hed and targets such as TNF-α,IL-6,and IL-1β.Experimentally,Hed was found to decrease RANKL,elevate osteoprotegerin(OPG),and suppress intestinalmRNA levels of pro-inflammatory cytokines such as IL-1β,IL-6,IL-17A,and TNF-α.Conclusion:Hed exerts significant anti-osteoporotic effects inOVX-induced osteoporosis through a dualmechanism involving the suppression of both osteoclastogenesis and innate immune signaling pathways.These findings highlighted Hed’s novel role in modulating immune-bone crosstalk,offering a promising strategy for treating osteolytic diseases without estrogenic side effects.
基金supported by project ZR2022MF330 supported by Shandong Provincial Natural Science Foundationthe National Natural Science Foundation of China under Grant No.61701286.
文摘Synthetic speech detection is an essential task in the field of voice security,aimed at identifying deceptive voice attacks generated by text-to-speech(TTS)systems or voice conversion(VC)systems.In this paper,we propose a synthetic speech detection model called TFTransformer,which integrates both local and global features to enhance detection capabilities by effectively modeling local and global dependencies.Structurally,the model is divided into two main components:a front-end and a back-end.The front-end of the model uses a combination of SincLayer and two-dimensional(2D)convolution to extract high-level feature maps(HFM)containing local dependency of the input speech signals.The back-end uses time-frequency Transformer module to process these feature maps and further capture global dependency.Furthermore,we propose TFTransformer-SE,which incorporates a channel attention mechanism within the 2D convolutional blocks.This enhancement aims to more effectively capture local dependencies,thereby improving the model’s performance.The experiments were conducted on the ASVspoof 2021 LA dataset,and the results showed that the model achieved an equal error rate(EER)of 3.37%without data augmentation.Additionally,we evaluated the model using the ASVspoof 2019 LA dataset,achieving an EER of 0.84%,also without data augmentation.This demonstrates that combining local and global dependencies in the time-frequency domain can significantly improve detection accuracy.
基金Project supported by the National Natural Science Foundation of China(Grant Nos.12104193 and U24A2017)National Undergraduate Training Program for Innovation and Entrepreneurship(Grant No.202411463037Z)the project of Changzhou Physics Society Fund(Grant No.CW20250102)。
文摘We theoretically investigate the phase sensitivity of a truncated SU(1,1)interferometer fed with a two-mode coherent state and employing double-port homodyne detection.On the one hand,we analytically demonstrate that the two-mode coherent state provides better phase sensitivity than the single-mode coherent state.In addition,we show that the doubleport homodyne detection is a quasi-optimal measurement.For a bright coherent-state input,the sensitivity of this scheme saturates the phase-sensitivity bound determined by the quantum Fisher information.On the other hand,we quantitatively illustrate the advantage of double-port homodyne detection over the single-port scheme under ideal conditions and in the presence of photon loss,respectively.Furthermore,our analysis indicates that the scheme we propose is robust against photon loss.
