Wearable wristband systems leverage deep learning to revolutionize hand gesture recognition in daily activities.Unlike existing approaches that often focus on static gestures and require extensive labeled data,the pro...Wearable wristband systems leverage deep learning to revolutionize hand gesture recognition in daily activities.Unlike existing approaches that often focus on static gestures and require extensive labeled data,the proposed wearable wristband with selfsupervised contrastive learning excels at dynamic motion tracking and adapts rapidly across multiple scenarios.It features a four-channel sensing array composed of an ionic hydrogel with hierarchical microcone structures and ultrathin flexible electrodes,resulting in high-sensitivity capacitance output.Through wireless transmission from a Wi-Fi module,the proposed algorithm learns latent features from the unlabeled signals of random wrist movements.Remarkably,only few-shot labeled data are sufficient for fine-tuning the model,enabling rapid adaptation to various tasks.The system achieves a high accuracy of 94.9%in different scenarios,including the prediction of eight-direction commands,and air-writing of all numbers and letters.The proposed method facilitates smooth transitions between multiple tasks without the need for modifying the structure or undergoing extensive task-specific training.Its utility has been further extended to enhance human–machine interaction over digital platforms,such as game controls,calculators,and three-language login systems,offering users a natural and intuitive way of communication.展开更多
The rapid integration of Internet of Things(IoT)technologies is reshaping the global energy landscape by deploying smart meters that enable high-resolution consumption monitoring,two-way communication,and advanced met...The rapid integration of Internet of Things(IoT)technologies is reshaping the global energy landscape by deploying smart meters that enable high-resolution consumption monitoring,two-way communication,and advanced metering infrastructure services.However,this digital transformation also exposes power system to evolving threats,ranging from cyber intrusions and electricity theft to device malfunctions,and the unpredictable nature of these anomalies,coupled with the scarcity of labeled fault data,makes realtime detection exceptionally challenging.To address these difficulties,a real-time decision support framework is presented for smart meter anomality detection that leverages rolling time windows and two self-supervised contrastive learning modules.The first module synthesizes diverse negative samples to overcome the lack of labeled anomalies,while the second captures intrinsic temporal patterns for enhanced contextual discrimination.The end-to-end framework continuously updates its model with rolling updated meter data to deliver timely identification of emerging abnormal behaviors in evolving grids.Extensive evaluations on eight publicly available smart meter datasets over seven diverse abnormal patterns testing demonstrate the effectiveness of the proposed full framework,achieving average recall and F1 score of more than 0.85.展开更多
BACKGROUND Gastrointestinal dual-contrast ultrasonography(DCUS)is characterized by its high resolution,sensitivity,and specificity.AIM To determine the accuracy of DCUS in predicting lymph node metastasis in middle-ag...BACKGROUND Gastrointestinal dual-contrast ultrasonography(DCUS)is characterized by its high resolution,sensitivity,and specificity.AIM To determine the accuracy of DCUS in predicting lymph node metastasis in middle-aged and elderly patients with gastric cancer(GC).METHODS A total of 100 middle-aged and elderly patients with GC admitted to the Fourth Affiliated Hospital of Soochow University(Dushu Lake Hospital,Suzhou,China)between April 2022 and April 2024 were selected.The baseline data and lymph node metastasis status were collected.DCUS combined with intravenous contrast technology was used to calculate the enhancement time(ET),time to peak(TTP),and slope of the ascending branch wash-in rate(WIR).These indicators were used in assessing lymph node metastasis in patients with GC.RESULTS Among 100 middle-aged and elderly patients with GC,35(35.00%)had lymph node metastases.GC patients with lymph node metastasis had a higher propor-tion of stage II TNM classification and higher WIR values than those without lymph node metastasis.The ET and TTP values were lower in patients with lymph node metastases,and all differences were statistically significant(P<0.05).The area under the curve values for ET,TTP,WIR,and combined diagnosis of GC lymph node metastasis using DCUS were all>0.7.Optimal assessment was achieved when the cutoff values for ET,TTP,and WIR were set at 16.32 seconds,10.67 seconds,and 7.02,res-pectively.CONCLUSION DCUS-mediated assessment of ET,TTP,and WIR can effectively predict and evaluate lymph node metastasis status in patients with GC,with higher sensitivity when used in combination.展开更多
BACKGROUND Oil-based iodinated contrast media have excellent contrast properties and are widely used for hysterosalpingographic evaluation of female infertility.On abdominal radiography and computed tomography(CT)scan...BACKGROUND Oil-based iodinated contrast media have excellent contrast properties and are widely used for hysterosalpingographic evaluation of female infertility.On abdominal radiography and computed tomography(CT)scans,their radiodensity is similar to that of metallic objects,which can sometimes lead to diagnostic confusion in the postoperative settings.In this case,retained oil-based contrast medium was observed on an abdominal radiograph following a cesarean section,making it difficult to differentiate from an intraperitoneal foreign body from surgery.The patient was a 37-year-old pregnant woman who was referred to our hospital at 32 weeks and 1 day of pregnancy due to complete placenta previa for mana-gement of pregnancy and delivery.An elective cesarean section was performed at 37 weeks and 3 days.A plain abdominal radiograph taken immediately after surgery revealed a near-round,hyperdense,mass-like shadow with a regular margin in the pelvic cavity.An intraperitoneal foreign body was suspected;therefore,an abdominal CT scan was performed.The foreign body was located on the left side of the pouch of Douglas and had a CT value of 7000 Hounsfield units,similar to that of metals.The CT value strongly suggested the presence of an artificial object.However,further inquiries with the patient and her previous physician revealed a history of hysterosalpingography.Accordingly,retained oil-based iodinated contrast medium was suspected,and observation of the object’s course was adopted.CONCLUSION When intraperitoneal foreign bodies are suspected on postoperative radiographs,the possibility of oil-based iodinated contrast medium retention should be considered.展开更多
Patent foramen ovale(PFO)is a common congenital heart disorder associated with stroke,decompression sickness and migraine.Combining synchronized contrast transcranial Doppler with contrast transthoracic echocardiograp...Patent foramen ovale(PFO)is a common congenital heart disorder associated with stroke,decompression sickness and migraine.Combining synchronized contrast transcranial Doppler with contrast transthoracic echocardiography has important clinical significance and can improve the accuracy of detecting right-left shunts(RLSs)in patients with PFO.In this letter,regarding an original study presented by Yao et al,we present our insights and discuss how to better help clinicians evaluate changes in PFO-related RLS.展开更多
AIM:To find the effective contrast enhancement method on retinal images for effective segmentation of retinal features.METHODS:A novel image preprocessing method that used neighbourhood-based improved contrast limited...AIM:To find the effective contrast enhancement method on retinal images for effective segmentation of retinal features.METHODS:A novel image preprocessing method that used neighbourhood-based improved contrast limited adaptive histogram equalization(NICLAHE)to improve retinal image contrast was suggested to aid in the accurate identification of retinal disorders and improve the visibility of fine retinal structures.Additionally,a minimal-order filter was applied to effectively denoise the images without compromising important retinal structures.The novel NICLAHE algorithm was inspired by the classical CLAHE algorithm,but enhanced it by selecting the clip limits and tile sized in a dynamical manner relative to the pixel values in an image as opposed to using fixed values.It was evaluated on the Drive and high-resolution fundus(HRF)datasets on conventional quality measures.RESULTS:The new proposed preprocessing technique was applied to two retinal image databases,Drive and HRF,with four quality metrics being,root mean square error(RMSE),peak signal to noise ratio(PSNR),root mean square contrast(RMSC),and overall contrast.The technique performed superiorly on both the data sets as compared to the traditional enhancement methods.In order to assess the compatibility of the method with automated diagnosis,a deep learning framework named ResNet was applied in the segmentation of retinal blood vessels.Sensitivity,specificity,precision and accuracy were used to analyse the performance.NICLAHE–enhanced images outperformed the traditional techniques on both the datasets with improved accuracy.CONCLUSION:NICLAHE provides better results than traditional methods with less error and improved contrastrelated values.These enhanced images are subsequently measured by sensitivity,specificity,precision,and accuracy,which yield a better result in both datasets.展开更多
Fisheye cameras offer a significantly larger field of view compared to conventional cameras,making them valuable tools in the field of computer vision.However,their unique optical characteristics often lead to image d...Fisheye cameras offer a significantly larger field of view compared to conventional cameras,making them valuable tools in the field of computer vision.However,their unique optical characteristics often lead to image distortions,which pose challenges for object detection tasks.To address this issue,we propose Yolo-CaSKA(Yolo with Contrastive Learning and Selective Kernel Attention),a novel training method that enhances object detection on fisheye camera images.The standard image and the corresponding distorted fisheye image pairs are used as positive samples,and the rest of the image pairs are used as negative samples,which are guided by contrastive learning to help the distorted images find the feature vectors of the corresponding normal images,to improve the detection accuracy.Additionally,we incorporate the Selective Kernel(SK)attention module to focus on regions prone to false detections,such as image edges and blind spots.Finally,the mAP_(50) on the augmented KITTI dataset is improved by 5.5% over the original Yolov8,while the mAP_(50) on the WoodScape dataset is improved by 2.6% compared to OmniDet.The results demonstrate the performance of our proposed model for object detection on fisheye images.展开更多
Neural machine translation(NMT)has advanced with deep learning and large-scale multilingual models,yet translating lowresource languages often lacks sufficient training data and leads to hallucinations.This often resu...Neural machine translation(NMT)has advanced with deep learning and large-scale multilingual models,yet translating lowresource languages often lacks sufficient training data and leads to hallucinations.This often results in translated content that diverges significantly from the source text.This research proposes a refined Contrastive Decoding(CD)algorithm that dynamically adjusts weights of log probabilities from strong expert and weak amateur models to mitigate hallucinations in lowresource NMT and improve translation quality.Advanced large language NMT models,including ChatGLM and LLaMA,are fine-tuned and implemented for their superior contextual understanding and cross-lingual capabilities.The refined CD algorithm evaluates multiple candidate translations using BLEU score,semantic similarity,and Named Entity Recognition accuracy.Extensive experimental results show substantial improvements in translation quality and a significant reduction in hallucination rates.Fine-tuned models achieve higher evaluation metrics compared to baseline models and state-of-the-art models.An ablation study confirms the contributions of each methodological component and highlights the effectiveness of the refined CD algorithm and advanced models in mitigating hallucinations.Notably,the refined methodology increased the BLEU score by approximately 30%compared to baseline models.展开更多
Sarcasm detection is a complex and challenging task,particularly in the context of Chinese social media,where it exhibits strong contextual dependencies and cultural specificity.To address the limitations of existing ...Sarcasm detection is a complex and challenging task,particularly in the context of Chinese social media,where it exhibits strong contextual dependencies and cultural specificity.To address the limitations of existing methods in capturing the implicit semantics and contextual associations in sarcastic expressions,this paper proposes an event-aware model for Chinese sarcasm detection,leveraging a multi-head attention(MHA)mechanism and contrastive learning(CL)strategies.The proposed model employs a dual-path Bidirectional Encoder Representations from Transformers(BERT)encoder to process comment text and event context separately and integrates an MHA mechanism to facilitate deep interactions between the two,thereby capturing multidimensional semantic associations.Additionally,a CL strategy is introduced to enhance feature representation capabilities,further improving the model’s performance in handling class imbalance and complex contextual scenarios.The model achieves state-of-the-art performance on the Chinese sarcasm dataset,with significant improvements in accuracy(79.55%),F1-score(84.22%),and an area under the curve(AUC,84.35%).展开更多
Emotion recognition plays a crucial role in various fields and is a key task in natural language processing (NLP). The objective is to identify and interpret emotional expressions in text. However, traditional emotion...Emotion recognition plays a crucial role in various fields and is a key task in natural language processing (NLP). The objective is to identify and interpret emotional expressions in text. However, traditional emotion recognition approaches often struggle in few-shot cross-domain scenarios due to their limited capacity to generalize semantic features across different domains. Additionally, these methods face challenges in accurately capturing complex emotional states, particularly those that are subtle or implicit. To overcome these limitations, we introduce a novel approach called Dual-Task Contrastive Meta-Learning (DTCML). This method combines meta-learning and contrastive learning to improve emotion recognition. Meta-learning enhances the model’s ability to generalize to new emotional tasks, while instance contrastive learning further refines the model by distinguishing unique features within each category, enabling it to better differentiate complex emotional expressions. Prototype contrastive learning, in turn, helps the model address the semantic complexity of emotions across different domains, enabling the model to learn fine-grained emotions expression. By leveraging dual tasks, DTCML learns from two domains simultaneously, the model is encouraged to learn more diverse and generalizable emotions features, thereby improving its cross-domain adaptability and robustness, and enhancing its generalization ability. We evaluated the performance of DTCML across four cross-domain settings, and the results show that our method outperforms the best baseline by 5.88%, 12.04%, 8.49%, and 8.40% in terms of accuracy.展开更多
Graph similarity learning aims to calculate the similarity between pairs of graphs.Existing unsupervised graph similarity learning methods based on contrastive learning encounter challenges related to random graph aug...Graph similarity learning aims to calculate the similarity between pairs of graphs.Existing unsupervised graph similarity learning methods based on contrastive learning encounter challenges related to random graph augmentation strategies,which can harm the semantic and structural information of graphs and overlook the rich structural information present in subgraphs.To address these issues,we propose a graph similarity learning model based on learnable augmentation and multi-level contrastive learning.First,to tackle the problem of random augmentation disrupting the semantics and structure of the graph,we design a learnable augmentation method to selectively choose nodes and edges within the graph.To enhance contrastive levels,we employ a biased random walk method to generate corresponding subgraphs,enriching the contrastive hierarchy.Second,to solve the issue of previous work not considering multi-level contrastive learning,we utilize graph convolutional networks to learn node representations of augmented views and the original graph and calculate the interaction information between the attribute-augmented and structure-augmented views and the original graph.The goal is to maximize node consistency between different views and learn node matching between different graphs,resulting in node-level representations for each graph.Subgraph representations are then obtained through pooling operations,and we conduct contrastive learning utilizing both node and subgraph representations.Finally,the graph similarity score is computed according to different downstream tasks.We conducted three sets of experiments across eight datasets,and the results demonstrate that the proposed model effectively mitigates the issues of random augmentation damaging the original graph’s semantics and structure,as well as the insufficiency of contrastive levels.Additionally,the model achieves the best overall performance.展开更多
Federated learning(FL)is a distributed machine learning paradigm for edge cloud computing.FL can facilitate data-driven decision-making in tactical scenarios,effectively addressing both data volume and infrastructure ...Federated learning(FL)is a distributed machine learning paradigm for edge cloud computing.FL can facilitate data-driven decision-making in tactical scenarios,effectively addressing both data volume and infrastructure challenges in edge environments.However,the diversity of clients in edge cloud computing presents significant challenges for FL.Personalized federated learning(pFL)received considerable attention in recent years.One example of pFL involves exploiting the global and local information in the local model.Current pFL algorithms experience limitations such as slow convergence speed,catastrophic forgetting,and poor performance in complex tasks,which still have significant shortcomings compared to the centralized learning.To achieve high pFL performance,we propose FedCLCC:Federated Contrastive Learning and Conditional Computing.The core of FedCLCC is the use of contrastive learning and conditional computing.Contrastive learning determines the feature representation similarity to adjust the local model.Conditional computing separates the global and local information and feeds it to their corresponding heads for global and local handling.Our comprehensive experiments demonstrate that FedCLCC outperforms other state-of-the-art FL algorithms.展开更多
Few-shot point cloud 3D object detection(FS3D)aims to identify and locate objects of novel classes within point clouds using knowledge acquired from annotated base classes and a minimal number of samples from the nove...Few-shot point cloud 3D object detection(FS3D)aims to identify and locate objects of novel classes within point clouds using knowledge acquired from annotated base classes and a minimal number of samples from the novel classes.Due to imbalanced training data,existing FS3D methods based on fully supervised learning can lead to overfitting toward base classes,which impairs the network’s ability to generalize knowledge learned from base classes to novel classes and also prevents the network from extracting distinctive foreground and background representations for novel class objects.To address these issues,this thesis proposes a category-agnostic contrastive learning approach,enhancing the generalization and identification abilities for almost unseen categories through the construction of pseudo-labels and positive-negative sample pairs unrelated to specific classes.Firstly,this thesis designs a proposal-wise context contrastive module(CCM).By reducing the distance between foreground point features and increasing the distance between foreground and background point features within a region proposal,CCM aids the network in extracting more discriminative foreground and background feature representations without reliance on categorical annotations.Secondly,this thesis utilizes a geometric contrastive module(GCM),which enhances the network’s geometric perception capability by employing contrastive learning on the foreground point features associated with various basic geometric components,such as edges,corners,and surfaces,thereby enabling these geometric components to exhibit more distinguishable representations.This thesis also combines category-aware contrastive learning with former modules to maintain categorical distinctiveness.Extensive experimental results on FS-SUNRGBD and FS-ScanNet datasets demonstrate the effectiveness of this method with average precision exceeding the baseline by up to 8%.展开更多
BACKGROUND Three-phase dynamic computed tomography imaging is particularly useful in the liver region.However,dynamic imaging with contrast media has the disadvantage of increased radiation exposure due to multiple im...BACKGROUND Three-phase dynamic computed tomography imaging is particularly useful in the liver region.However,dynamic imaging with contrast media has the disadvantage of increased radiation exposure due to multiple imaging sessions.We hypothesized that the contrast enhancement boost(CE-boost)technique could be used to enhance the contrast in equilibrium phase(EP)images and produce enhancement similar to that of portal vein phase(PVP)images,and if this is possible,EP imaging could play the same role as PVP imaging.We also speculated that this might allow the conversion of three-phase dynamic imaging to biphasic dynamic imaging,reducing patients’radiation exposure.AIM To determine if the CE-boost of EP,CE-boost(EP)is useful compared to a conventional image.METHODS We retrospectively analyzed the cases of 52 patients who were diagnosed with liver cancer between January 2016 and October 2022 at our institution.From these computed tomography images,CE-boost images were generated from the EP and plane images.We compared the PVP,EP,and CE-boost(EP)for blood vessels and hepatic parenchyma based on the contrast-to-noise ratio(CNR),signal-to-noise ratio,and figure-of-merit(FOM).Visual assessments were also performed for vessel visualization,lesion conspicuity,and image noise.RESULTS The CE-boost(EP)images showed significant superiority compared to the PVP images in the CNR,signal-to-noise ratio,and FOM except regarding the hepatic parenchyma.No significant differences were detected in CNR or FOM comparisons within the hepatic parenchyma(P=0.62,0.67).The comparison of the EP and CE-boost(EP)images consistently favored CE-boost(EP).Regarding the visual assessment,the CE-boost(EP)images were significantly superior to the PVP images in lesion conspicuity,and the PVP in image noise.The CE-boost(EP)images were significantly better than the EP images in the vessel visualization of segmental branches of the portal vein and lesion conspicuity,and the EP in image noise.CONCLUSION The image quality of CE-boost(EP)images was comparable or superior to that of conventional PVP and EP.CEboost(EP)images might provide information comparable to the conventional PVP.展开更多
Quantitative phase imaging(QPI)enables non-invasive cellular analysis by utilizing cell thickness and refractive index as intrinsic probes,revolutionizing label-free microscopy in cellular research.Differential phase ...Quantitative phase imaging(QPI)enables non-invasive cellular analysis by utilizing cell thickness and refractive index as intrinsic probes,revolutionizing label-free microscopy in cellular research.Differential phase contrast(DPC),a non-interferometric QPI technique,requires only four intensity images under asymmetric illumination to recover the phase of a sample,offering the advantages of being label-free,non-coherent and highly robust.Its phase reconstruction result relies on precise modeling of the phase transfer function(PTF).However,in real optical systems,the PTF will deviate from its theoretical ideal due to the unknown wavefront aberrations,which will lead to significant artifacts and distortions in the reconstructed phase.We propose an aberration-corrected DPC(ACDPC)method that utilizes three intensity images under annular illumination to jointly retrieve the aberration and the phase,achieving high-quality QPI with minimal raw data.By employing three annular illuminations precisely matched to the numerical aperture of the objective lens,the object information is transmitted into the acquired intensity with a high signal-to-noise ratio.Phase retrieval is achieved by an iterative deconvolution algorithm that uses simulated annealing to estimate the aberration and further employs regularized deconvolution to reconstruct the phase,ultimately obtaining a refined complex pupil function and an aberration-corrected quantitative phase.We demonstrate that ACDPC is robust to multi-order aberrations without any priori knowledge,and can effectively retrieve and correct system aberrations to obtain high-quality quantitative phase.Experimental results show that ACDPC can clearly reproduce subcellular structures such as vesicles and lipid droplets with higher resolution than conventional DPC,which opens up new possibilities for more accurate subcellular structure analysis in cell biology.展开更多
AIM:To investigate the effect of 0.01%low-concentration atropine(LA)on quantitative contrast sensitivity function(qCSF)in children with myopia.METHODS:This paired case-control study included 90 eyes of 58 children who...AIM:To investigate the effect of 0.01%low-concentration atropine(LA)on quantitative contrast sensitivity function(qCSF)in children with myopia.METHODS:This paired case-control study included 90 eyes of 58 children who were sex-,age-,and refractionmatched and equally divided into two groups:the 0.01%LA group had undergone 6mo use of daily 0.01%atropine and control group was naïve to LA.Routine ophthalmic examinations and qCSF test without refractive correction were performed.Two groups were compared in monocular and binocular qCSF parameters,including the area under logCSF,CSF acuity,and contrast sensitivity(CS)at 1.0-18.0 cycle per degree(cpd).RESULTS:In the monocular comparison,the CSF acuity of the LA group was significantly higher than that of the control group(7.58±5.51 vs 6.37±4.22 cpd,P<0.05).The subgroup analysis showed that in the 6-9y group,CSF acuity was significantly higher in the LA group than the control group(8.76±6.19 vs 6.54±4.25 cpd,P<0.05),and in the Female group,low refraction sphere group,and high refraction cylinder group,the CS at high spatial frequencies(12.0 and 18.0 cpd)were significantly higher in the LA group than in the control group(all P<0.05).In the binocular test,CSF acuity and CS at 12.0 cpd were significantly higher in the LA group than in the control group(10.95±7.00 vs 8.65±5.12 cpd;0.17±0.33 vs 0.06±0.16,respectively;both P<0.05).CONCLUSION:Use of LA may result in improved CS in children with early onset myopia.展开更多
AIM:To measure the contrast sensitivity(CS)using computer-based Chart2020 software pre-and post-white light exposure with and without blue-blocking lenses(BBLs).METHODS:The study included participants aged 18 to 25y(n...AIM:To measure the contrast sensitivity(CS)using computer-based Chart2020 software pre-and post-white light exposure with and without blue-blocking lenses(BBLs).METHODS:The study included participants aged 18 to 25y(n=30 eyes),where baseline CS was measured before the experiment.Following this,the participants were exposed to two white light-emitting diodes(LEDs;450 lx each),placed at a 45-degree angle from the participant’s eye and 80 cm from the light source.All participants were randomly divided into three groups(BBL1-Placebo lens,BBL2-Crizal Prevencia,BBL3-Duravision)by sequential randomisation,which was double-blinded.Post-light exposure,the CS was measured monocularly with a calibrated computer-based CS Chart-2020 software at different log units.RESULTS:CS measured using Chart-2020 software at 0.8,1.5,6,12,and 18 cpd pre-and post-white LED exposure with and without BBLs showed a significant difference(P<0.05)in contrast threshold and log contrast at 6 cpd and 18 cpd(P<0.05)and showed no significant differences in 0.8,1.5,12 cpd(P>0.05).CONCLUSION:This study shows that exposure to white LEDs can diminish CS,while BBLs may ameliorate these negative effects.展开更多
基金supported by the Research Grant Fund from Kwangwoon University in 2023,the National Natural Science Foundation of China under Grant(62311540155)the Taishan Scholars Project Special Funds(tsqn202312035)the open research foundation of State Key Laboratory of Integrated Chips and Systems.
文摘Wearable wristband systems leverage deep learning to revolutionize hand gesture recognition in daily activities.Unlike existing approaches that often focus on static gestures and require extensive labeled data,the proposed wearable wristband with selfsupervised contrastive learning excels at dynamic motion tracking and adapts rapidly across multiple scenarios.It features a four-channel sensing array composed of an ionic hydrogel with hierarchical microcone structures and ultrathin flexible electrodes,resulting in high-sensitivity capacitance output.Through wireless transmission from a Wi-Fi module,the proposed algorithm learns latent features from the unlabeled signals of random wrist movements.Remarkably,only few-shot labeled data are sufficient for fine-tuning the model,enabling rapid adaptation to various tasks.The system achieves a high accuracy of 94.9%in different scenarios,including the prediction of eight-direction commands,and air-writing of all numbers and letters.The proposed method facilitates smooth transitions between multiple tasks without the need for modifying the structure or undergoing extensive task-specific training.Its utility has been further extended to enhance human–machine interaction over digital platforms,such as game controls,calculators,and three-language login systems,offering users a natural and intuitive way of communication.
文摘The rapid integration of Internet of Things(IoT)technologies is reshaping the global energy landscape by deploying smart meters that enable high-resolution consumption monitoring,two-way communication,and advanced metering infrastructure services.However,this digital transformation also exposes power system to evolving threats,ranging from cyber intrusions and electricity theft to device malfunctions,and the unpredictable nature of these anomalies,coupled with the scarcity of labeled fault data,makes realtime detection exceptionally challenging.To address these difficulties,a real-time decision support framework is presented for smart meter anomality detection that leverages rolling time windows and two self-supervised contrastive learning modules.The first module synthesizes diverse negative samples to overcome the lack of labeled anomalies,while the second captures intrinsic temporal patterns for enhanced contextual discrimination.The end-to-end framework continuously updates its model with rolling updated meter data to deliver timely identification of emerging abnormal behaviors in evolving grids.Extensive evaluations on eight publicly available smart meter datasets over seven diverse abnormal patterns testing demonstrate the effectiveness of the proposed full framework,achieving average recall and F1 score of more than 0.85.
文摘BACKGROUND Gastrointestinal dual-contrast ultrasonography(DCUS)is characterized by its high resolution,sensitivity,and specificity.AIM To determine the accuracy of DCUS in predicting lymph node metastasis in middle-aged and elderly patients with gastric cancer(GC).METHODS A total of 100 middle-aged and elderly patients with GC admitted to the Fourth Affiliated Hospital of Soochow University(Dushu Lake Hospital,Suzhou,China)between April 2022 and April 2024 were selected.The baseline data and lymph node metastasis status were collected.DCUS combined with intravenous contrast technology was used to calculate the enhancement time(ET),time to peak(TTP),and slope of the ascending branch wash-in rate(WIR).These indicators were used in assessing lymph node metastasis in patients with GC.RESULTS Among 100 middle-aged and elderly patients with GC,35(35.00%)had lymph node metastases.GC patients with lymph node metastasis had a higher propor-tion of stage II TNM classification and higher WIR values than those without lymph node metastasis.The ET and TTP values were lower in patients with lymph node metastases,and all differences were statistically significant(P<0.05).The area under the curve values for ET,TTP,WIR,and combined diagnosis of GC lymph node metastasis using DCUS were all>0.7.Optimal assessment was achieved when the cutoff values for ET,TTP,and WIR were set at 16.32 seconds,10.67 seconds,and 7.02,res-pectively.CONCLUSION DCUS-mediated assessment of ET,TTP,and WIR can effectively predict and evaluate lymph node metastasis status in patients with GC,with higher sensitivity when used in combination.
文摘BACKGROUND Oil-based iodinated contrast media have excellent contrast properties and are widely used for hysterosalpingographic evaluation of female infertility.On abdominal radiography and computed tomography(CT)scans,their radiodensity is similar to that of metallic objects,which can sometimes lead to diagnostic confusion in the postoperative settings.In this case,retained oil-based contrast medium was observed on an abdominal radiograph following a cesarean section,making it difficult to differentiate from an intraperitoneal foreign body from surgery.The patient was a 37-year-old pregnant woman who was referred to our hospital at 32 weeks and 1 day of pregnancy due to complete placenta previa for mana-gement of pregnancy and delivery.An elective cesarean section was performed at 37 weeks and 3 days.A plain abdominal radiograph taken immediately after surgery revealed a near-round,hyperdense,mass-like shadow with a regular margin in the pelvic cavity.An intraperitoneal foreign body was suspected;therefore,an abdominal CT scan was performed.The foreign body was located on the left side of the pouch of Douglas and had a CT value of 7000 Hounsfield units,similar to that of metals.The CT value strongly suggested the presence of an artificial object.However,further inquiries with the patient and her previous physician revealed a history of hysterosalpingography.Accordingly,retained oil-based iodinated contrast medium was suspected,and observation of the object’s course was adopted.CONCLUSION When intraperitoneal foreign bodies are suspected on postoperative radiographs,the possibility of oil-based iodinated contrast medium retention should be considered.
文摘Patent foramen ovale(PFO)is a common congenital heart disorder associated with stroke,decompression sickness and migraine.Combining synchronized contrast transcranial Doppler with contrast transthoracic echocardiography has important clinical significance and can improve the accuracy of detecting right-left shunts(RLSs)in patients with PFO.In this letter,regarding an original study presented by Yao et al,we present our insights and discuss how to better help clinicians evaluate changes in PFO-related RLS.
文摘AIM:To find the effective contrast enhancement method on retinal images for effective segmentation of retinal features.METHODS:A novel image preprocessing method that used neighbourhood-based improved contrast limited adaptive histogram equalization(NICLAHE)to improve retinal image contrast was suggested to aid in the accurate identification of retinal disorders and improve the visibility of fine retinal structures.Additionally,a minimal-order filter was applied to effectively denoise the images without compromising important retinal structures.The novel NICLAHE algorithm was inspired by the classical CLAHE algorithm,but enhanced it by selecting the clip limits and tile sized in a dynamical manner relative to the pixel values in an image as opposed to using fixed values.It was evaluated on the Drive and high-resolution fundus(HRF)datasets on conventional quality measures.RESULTS:The new proposed preprocessing technique was applied to two retinal image databases,Drive and HRF,with four quality metrics being,root mean square error(RMSE),peak signal to noise ratio(PSNR),root mean square contrast(RMSC),and overall contrast.The technique performed superiorly on both the data sets as compared to the traditional enhancement methods.In order to assess the compatibility of the method with automated diagnosis,a deep learning framework named ResNet was applied in the segmentation of retinal blood vessels.Sensitivity,specificity,precision and accuracy were used to analyse the performance.NICLAHE–enhanced images outperformed the traditional techniques on both the datasets with improved accuracy.CONCLUSION:NICLAHE provides better results than traditional methods with less error and improved contrastrelated values.These enhanced images are subsequently measured by sensitivity,specificity,precision,and accuracy,which yield a better result in both datasets.
文摘Fisheye cameras offer a significantly larger field of view compared to conventional cameras,making them valuable tools in the field of computer vision.However,their unique optical characteristics often lead to image distortions,which pose challenges for object detection tasks.To address this issue,we propose Yolo-CaSKA(Yolo with Contrastive Learning and Selective Kernel Attention),a novel training method that enhances object detection on fisheye camera images.The standard image and the corresponding distorted fisheye image pairs are used as positive samples,and the rest of the image pairs are used as negative samples,which are guided by contrastive learning to help the distorted images find the feature vectors of the corresponding normal images,to improve the detection accuracy.Additionally,we incorporate the Selective Kernel(SK)attention module to focus on regions prone to false detections,such as image edges and blind spots.Finally,the mAP_(50) on the augmented KITTI dataset is improved by 5.5% over the original Yolov8,while the mAP_(50) on the WoodScape dataset is improved by 2.6% compared to OmniDet.The results demonstrate the performance of our proposed model for object detection on fisheye images.
基金M.Faheem is supported by VTT Technical Research Center of Finland.
文摘Neural machine translation(NMT)has advanced with deep learning and large-scale multilingual models,yet translating lowresource languages often lacks sufficient training data and leads to hallucinations.This often results in translated content that diverges significantly from the source text.This research proposes a refined Contrastive Decoding(CD)algorithm that dynamically adjusts weights of log probabilities from strong expert and weak amateur models to mitigate hallucinations in lowresource NMT and improve translation quality.Advanced large language NMT models,including ChatGLM and LLaMA,are fine-tuned and implemented for their superior contextual understanding and cross-lingual capabilities.The refined CD algorithm evaluates multiple candidate translations using BLEU score,semantic similarity,and Named Entity Recognition accuracy.Extensive experimental results show substantial improvements in translation quality and a significant reduction in hallucination rates.Fine-tuned models achieve higher evaluation metrics compared to baseline models and state-of-the-art models.An ablation study confirms the contributions of each methodological component and highlights the effectiveness of the refined CD algorithm and advanced models in mitigating hallucinations.Notably,the refined methodology increased the BLEU score by approximately 30%compared to baseline models.
基金granted by Qin Xin Talents Cultivation Program(No.QXTCP C202115),Beijing Information Science&Technology Universitythe Beijing Advanced Innovation Center for Future Blockchain and Privacy Computing Fund(No.GJJ-23),National Social Science Foundation,China(No.21BTQ079).
文摘Sarcasm detection is a complex and challenging task,particularly in the context of Chinese social media,where it exhibits strong contextual dependencies and cultural specificity.To address the limitations of existing methods in capturing the implicit semantics and contextual associations in sarcastic expressions,this paper proposes an event-aware model for Chinese sarcasm detection,leveraging a multi-head attention(MHA)mechanism and contrastive learning(CL)strategies.The proposed model employs a dual-path Bidirectional Encoder Representations from Transformers(BERT)encoder to process comment text and event context separately and integrates an MHA mechanism to facilitate deep interactions between the two,thereby capturing multidimensional semantic associations.Additionally,a CL strategy is introduced to enhance feature representation capabilities,further improving the model’s performance in handling class imbalance and complex contextual scenarios.The model achieves state-of-the-art performance on the Chinese sarcasm dataset,with significant improvements in accuracy(79.55%),F1-score(84.22%),and an area under the curve(AUC,84.35%).
基金supported by the ScientificResearch and Innovation Team Program of Sichuan University of Science and Technology(No.SUSE652A006)Sichuan Key Provincial Research Base of Intelligent Tourism(ZHYJ22-03)In addition,it is also listed as a project of Sichuan Provincial Science and Technology Programme(2022YFG0028).
文摘Emotion recognition plays a crucial role in various fields and is a key task in natural language processing (NLP). The objective is to identify and interpret emotional expressions in text. However, traditional emotion recognition approaches often struggle in few-shot cross-domain scenarios due to their limited capacity to generalize semantic features across different domains. Additionally, these methods face challenges in accurately capturing complex emotional states, particularly those that are subtle or implicit. To overcome these limitations, we introduce a novel approach called Dual-Task Contrastive Meta-Learning (DTCML). This method combines meta-learning and contrastive learning to improve emotion recognition. Meta-learning enhances the model’s ability to generalize to new emotional tasks, while instance contrastive learning further refines the model by distinguishing unique features within each category, enabling it to better differentiate complex emotional expressions. Prototype contrastive learning, in turn, helps the model address the semantic complexity of emotions across different domains, enabling the model to learn fine-grained emotions expression. By leveraging dual tasks, DTCML learns from two domains simultaneously, the model is encouraged to learn more diverse and generalizable emotions features, thereby improving its cross-domain adaptability and robustness, and enhancing its generalization ability. We evaluated the performance of DTCML across four cross-domain settings, and the results show that our method outperforms the best baseline by 5.88%, 12.04%, 8.49%, and 8.40% in terms of accuracy.
文摘Graph similarity learning aims to calculate the similarity between pairs of graphs.Existing unsupervised graph similarity learning methods based on contrastive learning encounter challenges related to random graph augmentation strategies,which can harm the semantic and structural information of graphs and overlook the rich structural information present in subgraphs.To address these issues,we propose a graph similarity learning model based on learnable augmentation and multi-level contrastive learning.First,to tackle the problem of random augmentation disrupting the semantics and structure of the graph,we design a learnable augmentation method to selectively choose nodes and edges within the graph.To enhance contrastive levels,we employ a biased random walk method to generate corresponding subgraphs,enriching the contrastive hierarchy.Second,to solve the issue of previous work not considering multi-level contrastive learning,we utilize graph convolutional networks to learn node representations of augmented views and the original graph and calculate the interaction information between the attribute-augmented and structure-augmented views and the original graph.The goal is to maximize node consistency between different views and learn node matching between different graphs,resulting in node-level representations for each graph.Subgraph representations are then obtained through pooling operations,and we conduct contrastive learning utilizing both node and subgraph representations.Finally,the graph similarity score is computed according to different downstream tasks.We conducted three sets of experiments across eight datasets,and the results demonstrate that the proposed model effectively mitigates the issues of random augmentation damaging the original graph’s semantics and structure,as well as the insufficiency of contrastive levels.Additionally,the model achieves the best overall performance.
基金supported by the Natural Science Foundation of Xinjiang Uygur Autonomous Region(Grant No.2022D01B 187)。
文摘Federated learning(FL)is a distributed machine learning paradigm for edge cloud computing.FL can facilitate data-driven decision-making in tactical scenarios,effectively addressing both data volume and infrastructure challenges in edge environments.However,the diversity of clients in edge cloud computing presents significant challenges for FL.Personalized federated learning(pFL)received considerable attention in recent years.One example of pFL involves exploiting the global and local information in the local model.Current pFL algorithms experience limitations such as slow convergence speed,catastrophic forgetting,and poor performance in complex tasks,which still have significant shortcomings compared to the centralized learning.To achieve high pFL performance,we propose FedCLCC:Federated Contrastive Learning and Conditional Computing.The core of FedCLCC is the use of contrastive learning and conditional computing.Contrastive learning determines the feature representation similarity to adjust the local model.Conditional computing separates the global and local information and feeds it to their corresponding heads for global and local handling.Our comprehensive experiments demonstrate that FedCLCC outperforms other state-of-the-art FL algorithms.
文摘Few-shot point cloud 3D object detection(FS3D)aims to identify and locate objects of novel classes within point clouds using knowledge acquired from annotated base classes and a minimal number of samples from the novel classes.Due to imbalanced training data,existing FS3D methods based on fully supervised learning can lead to overfitting toward base classes,which impairs the network’s ability to generalize knowledge learned from base classes to novel classes and also prevents the network from extracting distinctive foreground and background representations for novel class objects.To address these issues,this thesis proposes a category-agnostic contrastive learning approach,enhancing the generalization and identification abilities for almost unseen categories through the construction of pseudo-labels and positive-negative sample pairs unrelated to specific classes.Firstly,this thesis designs a proposal-wise context contrastive module(CCM).By reducing the distance between foreground point features and increasing the distance between foreground and background point features within a region proposal,CCM aids the network in extracting more discriminative foreground and background feature representations without reliance on categorical annotations.Secondly,this thesis utilizes a geometric contrastive module(GCM),which enhances the network’s geometric perception capability by employing contrastive learning on the foreground point features associated with various basic geometric components,such as edges,corners,and surfaces,thereby enabling these geometric components to exhibit more distinguishable representations.This thesis also combines category-aware contrastive learning with former modules to maintain categorical distinctiveness.Extensive experimental results on FS-SUNRGBD and FS-ScanNet datasets demonstrate the effectiveness of this method with average precision exceeding the baseline by up to 8%.
文摘BACKGROUND Three-phase dynamic computed tomography imaging is particularly useful in the liver region.However,dynamic imaging with contrast media has the disadvantage of increased radiation exposure due to multiple imaging sessions.We hypothesized that the contrast enhancement boost(CE-boost)technique could be used to enhance the contrast in equilibrium phase(EP)images and produce enhancement similar to that of portal vein phase(PVP)images,and if this is possible,EP imaging could play the same role as PVP imaging.We also speculated that this might allow the conversion of three-phase dynamic imaging to biphasic dynamic imaging,reducing patients’radiation exposure.AIM To determine if the CE-boost of EP,CE-boost(EP)is useful compared to a conventional image.METHODS We retrospectively analyzed the cases of 52 patients who were diagnosed with liver cancer between January 2016 and October 2022 at our institution.From these computed tomography images,CE-boost images were generated from the EP and plane images.We compared the PVP,EP,and CE-boost(EP)for blood vessels and hepatic parenchyma based on the contrast-to-noise ratio(CNR),signal-to-noise ratio,and figure-of-merit(FOM).Visual assessments were also performed for vessel visualization,lesion conspicuity,and image noise.RESULTS The CE-boost(EP)images showed significant superiority compared to the PVP images in the CNR,signal-to-noise ratio,and FOM except regarding the hepatic parenchyma.No significant differences were detected in CNR or FOM comparisons within the hepatic parenchyma(P=0.62,0.67).The comparison of the EP and CE-boost(EP)images consistently favored CE-boost(EP).Regarding the visual assessment,the CE-boost(EP)images were significantly superior to the PVP images in lesion conspicuity,and the PVP in image noise.The CE-boost(EP)images were significantly better than the EP images in the vessel visualization of segmental branches of the portal vein and lesion conspicuity,and the EP in image noise.CONCLUSION The image quality of CE-boost(EP)images was comparable or superior to that of conventional PVP and EP.CEboost(EP)images might provide information comparable to the conventional PVP.
基金supported by the National Natural Science Foundation of China(62305162,62227818,62361136588)China Postdoctoral Science Foundation(2023TQ0160,2023M731683)+5 种基金Nanjing University of Science and Technology independent research project(30923010305)National Key Research and Development Program of China(2024YFE0101300)Biomedical Competition Foundation of Jiangsu Province(BE2022847)Key National Industrial Technology Cooperation Foundation of Jiangsu Province(BZ2022039)Fundamental Research Funds for the Central Universities(2023102001)Open Research Fund of Jiangsu Key Laboratory of Spectral Imaging&Intelligent Sense(JSGP202105,JSGP202201,JSGPCXZNGZ202401)。
文摘Quantitative phase imaging(QPI)enables non-invasive cellular analysis by utilizing cell thickness and refractive index as intrinsic probes,revolutionizing label-free microscopy in cellular research.Differential phase contrast(DPC),a non-interferometric QPI technique,requires only four intensity images under asymmetric illumination to recover the phase of a sample,offering the advantages of being label-free,non-coherent and highly robust.Its phase reconstruction result relies on precise modeling of the phase transfer function(PTF).However,in real optical systems,the PTF will deviate from its theoretical ideal due to the unknown wavefront aberrations,which will lead to significant artifacts and distortions in the reconstructed phase.We propose an aberration-corrected DPC(ACDPC)method that utilizes three intensity images under annular illumination to jointly retrieve the aberration and the phase,achieving high-quality QPI with minimal raw data.By employing three annular illuminations precisely matched to the numerical aperture of the objective lens,the object information is transmitted into the acquired intensity with a high signal-to-noise ratio.Phase retrieval is achieved by an iterative deconvolution algorithm that uses simulated annealing to estimate the aberration and further employs regularized deconvolution to reconstruct the phase,ultimately obtaining a refined complex pupil function and an aberration-corrected quantitative phase.We demonstrate that ACDPC is robust to multi-order aberrations without any priori knowledge,and can effectively retrieve and correct system aberrations to obtain high-quality quantitative phase.Experimental results show that ACDPC can clearly reproduce subcellular structures such as vesicles and lipid droplets with higher resolution than conventional DPC,which opens up new possibilities for more accurate subcellular structure analysis in cell biology.
基金Supported by National Key Research and Development Program of China(No.2023YFA0915000)。
文摘AIM:To investigate the effect of 0.01%low-concentration atropine(LA)on quantitative contrast sensitivity function(qCSF)in children with myopia.METHODS:This paired case-control study included 90 eyes of 58 children who were sex-,age-,and refractionmatched and equally divided into two groups:the 0.01%LA group had undergone 6mo use of daily 0.01%atropine and control group was naïve to LA.Routine ophthalmic examinations and qCSF test without refractive correction were performed.Two groups were compared in monocular and binocular qCSF parameters,including the area under logCSF,CSF acuity,and contrast sensitivity(CS)at 1.0-18.0 cycle per degree(cpd).RESULTS:In the monocular comparison,the CSF acuity of the LA group was significantly higher than that of the control group(7.58±5.51 vs 6.37±4.22 cpd,P<0.05).The subgroup analysis showed that in the 6-9y group,CSF acuity was significantly higher in the LA group than the control group(8.76±6.19 vs 6.54±4.25 cpd,P<0.05),and in the Female group,low refraction sphere group,and high refraction cylinder group,the CS at high spatial frequencies(12.0 and 18.0 cpd)were significantly higher in the LA group than in the control group(all P<0.05).In the binocular test,CSF acuity and CS at 12.0 cpd were significantly higher in the LA group than in the control group(10.95±7.00 vs 8.65±5.12 cpd;0.17±0.33 vs 0.06±0.16,respectively;both P<0.05).CONCLUSION:Use of LA may result in improved CS in children with early onset myopia.
文摘AIM:To measure the contrast sensitivity(CS)using computer-based Chart2020 software pre-and post-white light exposure with and without blue-blocking lenses(BBLs).METHODS:The study included participants aged 18 to 25y(n=30 eyes),where baseline CS was measured before the experiment.Following this,the participants were exposed to two white light-emitting diodes(LEDs;450 lx each),placed at a 45-degree angle from the participant’s eye and 80 cm from the light source.All participants were randomly divided into three groups(BBL1-Placebo lens,BBL2-Crizal Prevencia,BBL3-Duravision)by sequential randomisation,which was double-blinded.Post-light exposure,the CS was measured monocularly with a calibrated computer-based CS Chart-2020 software at different log units.RESULTS:CS measured using Chart-2020 software at 0.8,1.5,6,12,and 18 cpd pre-and post-white LED exposure with and without BBLs showed a significant difference(P<0.05)in contrast threshold and log contrast at 6 cpd and 18 cpd(P<0.05)and showed no significant differences in 0.8,1.5,12 cpd(P>0.05).CONCLUSION:This study shows that exposure to white LEDs can diminish CS,while BBLs may ameliorate these negative effects.
