As healthcare systems increasingly embrace digitalization,effective management of electronic health records(EHRs)has emerged as a critical priority,particularly in inpatient settings where data sensitivity and realtim...As healthcare systems increasingly embrace digitalization,effective management of electronic health records(EHRs)has emerged as a critical priority,particularly in inpatient settings where data sensitivity and realtime access are paramount.Traditional EHR systems face significant challenges,including unauthorized access,data breaches,and inefficiencies in tracking follow-up appointments,which heighten the risk of misdiagnosis and medication errors.To address these issues,this research proposes a hybrid blockchain-based solution for securely managing EHRs,specifically designed as a framework for tracking inpatient follow-ups.By integrating QR codeenabled data access with a blockchain architecture,this innovative approach enhances privacy protection,data integrity,and auditing capabilities,while facilitating swift and real-time data retrieval.The architecture adheres to Role-Based Access Control(RBAC)principles and utilizes robust encryption techniques,including SHA-256 and AES-256-CBC,to secure sensitive information.A comprehensive threat model outlines trust boundaries and potential adversaries,complemented by a validated data transmission protocol.Experimental results demonstrate that the framework remains reliable in concurrent access scenarios,highlighting its efficiency and responsiveness in real-world applications.This study emphasizes the necessity for hybrid solutions in managing sensitive medical information and advocates for integrating blockchain technology and QR code innovations into contemporary healthcare systems.展开更多
Objective:To explore the impact of systematic stepwise rehabilitation nursing intervention on the prognosis and disease uncertainty of patients with hypertensive intracerebral hemorrhage,and to provide feasible strate...Objective:To explore the impact of systematic stepwise rehabilitation nursing intervention on the prognosis and disease uncertainty of patients with hypertensive intracerebral hemorrhage,and to provide feasible strategies for clinical nursing.Methods:Eighty patients with hypertensive intracerebral hemorrhage admitted to our hospital from January 2023 to June 2025 were selected and randomly divided into an observation group(n=40,receiving systematic stepwise rehabilitation nursing)and a control group(n=40,receiving conventional nursing).The intervention effects were analyzed by comparing changes in the National Institutes of Health Stroke Scale(NIHSS)scores for neurological recovery,Short Form 36 Health Survey(SF-36)scores for quality of life,Exercise of Self-Care Agency Scale(ESCA)scores for self-management ability,compliance,and the Mishel Uncertainty in Illness Scale(MUIS)scores between the two groups.Results:All scores in the observation group were significantly better than those in the control group after the intervention(p<0.05).Specifically,the NIHSS scores decreased more significantly,the total SF-36 scores increased,the ESCA scores increased significantly,while the MUIS scores decreased significantly,and compliance improved markedly,indicating a reduction in disease uncertainty among patients.Conclusion:Systematic stepwise rehabilitation nursing intervention can significantly improve neurological recovery,quality of life,self-management ability,and compliance in patients with hypertensive intracerebral hemorrhage,while effectively reducing disease uncertainty.It is worthy of clinical promotion and application.展开更多
If you look into the EU AI Act,you will find that the requirements are specified at a very high level.This is because the details are defined elsewhere.That is what standardization is doing.If we look at the world of ...If you look into the EU AI Act,you will find that the requirements are specified at a very high level.This is because the details are defined elsewhere.That is what standardization is doing.If we look at the world of AI standardization,there are three tiers-the international tier,the European tier and the national tier.There are also AI standardization committees at all three levels.展开更多
At present,the AI field is in a golden window period,which provides a historic opportunity for establishing new AI standards.Over the years,sensing enhanced AI technology has constantly developed,promoting the develop...At present,the AI field is in a golden window period,which provides a historic opportunity for establishing new AI standards.Over the years,sensing enhanced AI technology has constantly developed,promoting the development and progress of industries.With the development of new technologies,the demand for standards has become more prominent.The Global Center for Sensing Enhanced AI(GCSEA)is expected to integrate domestic and foreign resources and make joint efforts to promote the development of relevant sensing standards.展开更多
Internal solitary waves(ISWs)are an essential dynamic process in the ocean due to their large amplitude and long propagation distance.Traditional satellite observations provide only twodimensional observations of ocea...Internal solitary waves(ISWs)are an essential dynamic process in the ocean due to their large amplitude and long propagation distance.Traditional satellite observations provide only twodimensional observations of ocean signatures induced by ISWs.The Surface Water and Ocean Topography(SWOT)satellite has drawn significant attention due to its high resolution and threedimensional observation capabilities.SWOT can generate high-precision three-dimensional sea surface topography,capture sea surface undulations,and reveal ISW-related surface oscillations,thus offering a new perspective for studying ISWs.We collected 43 SWOT observations with clear ISW signatures in the Lombok Strait from August 2023 to June 2024.Based on collected data,the ISW imaging characteristics and distributions were analyzed,and the ISW-related sea level anomaly(SLA)data were measured by the SWOT to calculate the ISW amplitude and reveal the amplitude variations during the propagation along the wave crest.The ISW amplitudes generally range between 10 and 100 m,with most ISW amplitudes between 20 and 40 m.By analyzing two consecutive generated ISW packets,we identified the spreading effect along ISW wave crests,which manifests as ISW amplitude decrease with increase in propagation distance,and the amplitude distribution is non-uniform along the wave crest.Further analysis of the propagation paths of the maximum amplitude of ISW moving northward through the Lombok Strait revealed that these maxima are predominantly oriented in northeast direction.Finally,the relationship between the amplitude of ISW and the resulting SLA was analyzed.The Pearson correlation coefficient between these two variables is as high as 0.90,which suggests a strong positive correlation between amplitude and SLA.Furthermore,this relationship is closely related to the water depth,indicating that the three-dimensional sea surface observations provided by SWOT offer crucial observational data for the inversion of amplitudes of ISW.展开更多
ropic 1:Regarding sustainable development and global public interests,what should international Al standards focus on?James Ong:Since 2019,I have witnessed the evolution of WAIC and found that a consensus on the philo...ropic 1:Regarding sustainable development and global public interests,what should international Al standards focus on?James Ong:Since 2019,I have witnessed the evolution of WAIC and found that a consensus on the philosophical and ethic level on advocating“AI for humanity”is necessary,since ethics factor carries more weight in standards development.I want to emphasize three points:AI assisting sustainable development,AI empowering a balanced global development,and human-AI coordination for preventing AI risks.展开更多
Gut-brain communication via the peripheral neural network is vital for regulating local digestive function and systemic physiology.Gut microbiota,which produces a wide array of neuroactive compounds,is a critical modu...Gut-brain communication via the peripheral neural network is vital for regulating local digestive function and systemic physiology.Gut microbiota,which produces a wide array of neuroactive compounds,is a critical modulator in this bidirectional dialog.Perturbations in the gut microbiota have been implicated in neurological disorders such as depression and stress.Distinct from humans and other monogastric animals,ruminants possess a unique,microbially dense gastrointestinal compartment,the rumen,that facilitates the digestion of fibrous plant materials.These ruminal microbes are likely key contributors to rumen-brain crosstalk.Unlike certain microbe-derived neuroactive compounds produced in the hindgut that are minimally absorbed and primarily excreted in feces,those generated in rumen can reach the small intestine,where they are largely absorbed and affect central nervous system through systemic regulation in addition to the vagal pathway.Notably,emerging evidence suggests that rumen microbiota dysbiosis under stress is associated with abnormal behavior,altered hormonal and neurotransmitter levels.In this review,we introduce the concept of the rumen-microbiome-brain axis by comparing the anatomical structures and microbial characteristics of the intestine and the rumen,emphasizing the neuroactive potential of rumen microbiome and underlying mechanisms.Advances in this frontier hold tremendous promise to reveal a novel dimension of the gut-microbiome-brain axis,providing transformative opportunities to improve ruminant welfare,productivity,and agricultural sustainability.展开更多
Robotic computing systems play an important role in enabling intelligent robotic tasks through intelligent algo-rithms and supporting hardware.In recent years,the evolution of robotic algorithms indicates a roadmap fr...Robotic computing systems play an important role in enabling intelligent robotic tasks through intelligent algo-rithms and supporting hardware.In recent years,the evolution of robotic algorithms indicates a roadmap from traditional robotics to hierarchical and end-to-end models.This algorithmic advancement poses a critical challenge in achieving balanced system-wide performance.Therefore,algorithm-hardware co-design has emerged as the primary methodology,which ana-lyzes algorithm behaviors on hardware to identify common computational properties.These properties can motivate algo-rithm optimization to reduce computational complexity and hardware innovation from architecture to circuit for high performance and high energy efficiency.We then reviewed recent works on robotic and embodied AI algorithms and computing hard-ware to demonstrate this algorithm-hardware co-design methodology.In the end,we discuss future research opportunities by answering two questions:(1)how to adapt the computing platforms to the rapid evolution of embodied AI algorithms,and(2)how to transform the potential of emerging hardware innovations into end-to-end inference improvements.展开更多
Heart disease remains a leading cause of mortality worldwide,emphasizing the urgent need for reliable and interpretable predictive models to support early diagnosis and timely intervention.However,existing Deep Learni...Heart disease remains a leading cause of mortality worldwide,emphasizing the urgent need for reliable and interpretable predictive models to support early diagnosis and timely intervention.However,existing Deep Learning(DL)approaches often face several limitations,including inefficient feature extraction,class imbalance,suboptimal classification performance,and limited interpretability,which collectively hinder their deployment in clinical settings.To address these challenges,we propose a novel DL framework for heart disease prediction that integrates a comprehensive preprocessing pipeline with an advanced classification architecture.The preprocessing stage involves label encoding and feature scaling.To address the issue of class imbalance inherent in the personal key indicators of the heart disease dataset,the localized random affine shadowsampling technique is employed,which enhances minority class representation while minimizing overfitting.At the core of the framework lies the Deep Residual Network(DeepResNet),which employs hierarchical residual transformations to facilitate efficient feature extraction and capture complex,non-linear relationships in the data.Experimental results demonstrate that the proposed model significantly outperforms existing techniques,achieving improvements of 3.26%in accuracy,3.16%in area under the receiver operating characteristics,1.09%in recall,and 1.07%in F1-score.Furthermore,robustness is validated using 10-fold crossvalidation,confirming the model’s generalizability across diverse data distributions.Moreover,model interpretability is ensured through the integration of Shapley additive explanations and local interpretable model-agnostic explanations,offering valuable insights into the contribution of individual features to model predictions.Overall,the proposed DL framework presents a robust,interpretable,and clinically applicable solution for heart disease prediction.展开更多
Manual inspection of onba earing casting defects is not realistic and unreliable,particularly in the case of some micro-level anomalies which lead to major defects on a large scale.To address these challenges,we propo...Manual inspection of onba earing casting defects is not realistic and unreliable,particularly in the case of some micro-level anomalies which lead to major defects on a large scale.To address these challenges,we propose BearFusionNet,an attention-based deep learning architecture with multi-stream,which merges both DenseNet201 and MobileNetV2 for feature extraction with a classification head inspired by VGG19.This hybrid design,figuratively beaming from one layer to another,extracts the enormity of representations on different scales,backed by a prepreprocessing pipeline that brings defect saliency to the fore through contrast adjustment,denoising,and edge detection.The use of multi-head self-attention enhances feature fusion,enabling the model to capture both large and small spatial features.BearFusionNet achieves an accuracy of 99.66%and Cohen’s kappa score of 0.9929 in Kaggle’s Real-life Industrial Casting Defects dataset.Both McNemar’s and Wilcoxon signed-rank statistical tests,as well as fivefold cross-validation,are employed to assess the robustness of our proposed model.To interpret the model,we adopt Grad-Cam visualizations,which are the state of the art standard.Furthermore,we deploy BearFusionNet as a webbased system for near real-time inference(5-6 s per prediction),which enables the quickest yet accurate detection with visual explanations.Overall,BearFusionNet is an interpretable,accurate,and deployable solution that can automatically detect casting defects,leading to significant advances in the innovative industrial environment.展开更多
It’s a pleasure to be here and speak about the industrial copilot and generative AI and the changing applications.As you all know,generative AI has arrived since a few years ago,we have generative AI not only in the ...It’s a pleasure to be here and speak about the industrial copilot and generative AI and the changing applications.As you all know,generative AI has arrived since a few years ago,we have generative AI not only in the consumer world,but also in the industrial world.Siemens is very active in the industrial space.We need to make AI real,because at the end in industry manufacturing,you are producing parts in the real world.So we need to make sure that AI and its applications can interact and comply with the real world.展开更多
Poststro ke cognitive impairment is a major secondary effect of ischemic stroke in many patients;however,few options are available for the early diagnosis and treatment of this condition.The aims of this study were to...Poststro ke cognitive impairment is a major secondary effect of ischemic stroke in many patients;however,few options are available for the early diagnosis and treatment of this condition.The aims of this study were to(1)determine the specific relationship between hypoxic andα-synuclein during the occur of poststroke cognitive impairment and(2)assess whether the serum phosphorylatedα-synuclein level can be used as a biomarker for poststro ke cognitive impairment.We found that the phosphorylatedα-synuclein level was significantly increased and showed pathological aggregation around the cerebral infa rct area in a mouse model of ischemic stroke.In addition,neuronalα-synuclein phosphorylation and aggregation were observed in the brain tissue of mice subjected to chronic hypoxia,suggesting that hypoxia is the underlying cause ofα-synuclein-mediated pathology in the brains of mice with ischemic stroke.Serum phosphorylatedα-synuclein levels in patients with ischemic stroke were significantly lower than those in healt hy subjects,and were positively correlated with cognition levels in patients with ischemic stroke.Furthermore,a decrease in serum high-density lipoprotein levels in stroke patie nts was significantly correlated with a decrease in phosphorylatedα-synuclein levels.Although ischemic stroke mice did not show significant cognitive impairment or disrupted lipid metabolism 14 days after injury,some of them exhibited decreased cognitive function and reduced phosphorylatedα-synuclein levels.Taken together,our results suggest that serum phosphorylatedα-synuclein is a potential biomarker for poststroke cognitive impairment.展开更多
To avoid the laborious annotation process for dense prediction tasks like semantic segmentation,unsupervised domain adaptation(UDA)methods have been proposed to leverage the abundant annotations from a source domain,s...To avoid the laborious annotation process for dense prediction tasks like semantic segmentation,unsupervised domain adaptation(UDA)methods have been proposed to leverage the abundant annotations from a source domain,such as virtual world(e.g.,3D games),and adapt models to the target domain(the real world)by narrowing the domain discrepancies.However,because of the large domain gap,directly aligning two distinct domains without considering the intermediates leads to inefficient alignment and inferior adaptation.To address this issue,we propose a novel learnable evolutionary Category Intermediates(CIs)guided UDA model named Leci,which enables the information transfer between the two domains via two processes,i.e.,Distilling and Blending.Starting from a random initialization,the CIs learn shared category-wise semantics automatically from two domains in the Distilling process.Then,the learned semantics in the CIs are sent back to blend the domain features through a residual attentive fusion(RAF)module,such that the categorywise features of both domains shift towards each other.As the CIs progressively and consistently learn from the varying feature distributions during training,they are evolutionary to guide the model to achieve category-wise feature alignment.Experiments on both GTA5 and SYNTHIA datasets demonstrate Leci's superiority over prior representative methods.展开更多
基金funded by Multimedia University,Cyberjaya,Selangor,Malaysia(Grant Number:PostDoc(MMUI/240029)).
文摘As healthcare systems increasingly embrace digitalization,effective management of electronic health records(EHRs)has emerged as a critical priority,particularly in inpatient settings where data sensitivity and realtime access are paramount.Traditional EHR systems face significant challenges,including unauthorized access,data breaches,and inefficiencies in tracking follow-up appointments,which heighten the risk of misdiagnosis and medication errors.To address these issues,this research proposes a hybrid blockchain-based solution for securely managing EHRs,specifically designed as a framework for tracking inpatient follow-ups.By integrating QR codeenabled data access with a blockchain architecture,this innovative approach enhances privacy protection,data integrity,and auditing capabilities,while facilitating swift and real-time data retrieval.The architecture adheres to Role-Based Access Control(RBAC)principles and utilizes robust encryption techniques,including SHA-256 and AES-256-CBC,to secure sensitive information.A comprehensive threat model outlines trust boundaries and potential adversaries,complemented by a validated data transmission protocol.Experimental results demonstrate that the framework remains reliable in concurrent access scenarios,highlighting its efficiency and responsiveness in real-world applications.This study emphasizes the necessity for hybrid solutions in managing sensitive medical information and advocates for integrating blockchain technology and QR code innovations into contemporary healthcare systems.
文摘Objective:To explore the impact of systematic stepwise rehabilitation nursing intervention on the prognosis and disease uncertainty of patients with hypertensive intracerebral hemorrhage,and to provide feasible strategies for clinical nursing.Methods:Eighty patients with hypertensive intracerebral hemorrhage admitted to our hospital from January 2023 to June 2025 were selected and randomly divided into an observation group(n=40,receiving systematic stepwise rehabilitation nursing)and a control group(n=40,receiving conventional nursing).The intervention effects were analyzed by comparing changes in the National Institutes of Health Stroke Scale(NIHSS)scores for neurological recovery,Short Form 36 Health Survey(SF-36)scores for quality of life,Exercise of Self-Care Agency Scale(ESCA)scores for self-management ability,compliance,and the Mishel Uncertainty in Illness Scale(MUIS)scores between the two groups.Results:All scores in the observation group were significantly better than those in the control group after the intervention(p<0.05).Specifically,the NIHSS scores decreased more significantly,the total SF-36 scores increased,the ESCA scores increased significantly,while the MUIS scores decreased significantly,and compliance improved markedly,indicating a reduction in disease uncertainty among patients.Conclusion:Systematic stepwise rehabilitation nursing intervention can significantly improve neurological recovery,quality of life,self-management ability,and compliance in patients with hypertensive intracerebral hemorrhage,while effectively reducing disease uncertainty.It is worthy of clinical promotion and application.
文摘If you look into the EU AI Act,you will find that the requirements are specified at a very high level.This is because the details are defined elsewhere.That is what standardization is doing.If we look at the world of AI standardization,there are three tiers-the international tier,the European tier and the national tier.There are also AI standardization committees at all three levels.
文摘At present,the AI field is in a golden window period,which provides a historic opportunity for establishing new AI standards.Over the years,sensing enhanced AI technology has constantly developed,promoting the development and progress of industries.With the development of new technologies,the demand for standards has become more prominent.The Global Center for Sensing Enhanced AI(GCSEA)is expected to integrate domestic and foreign resources and make joint efforts to promote the development of relevant sensing standards.
基金Supported by the National Key Research and Development Program of China(No.2021YFB3901304)the Shandong Provincial Natural Science Foundation(No.ZR2024QD054)+2 种基金the National Key Research and Development Program of China(No.2019YFA0606702)the National Natural Science Foundation of China(Nos.41906157,42306194,42306195)the Oceanographic Data Center,Chinese Academy of Sciences and the platform of Sino-Indonesian Joint Laboratory for Marine Sciences(SIMS)。
文摘Internal solitary waves(ISWs)are an essential dynamic process in the ocean due to their large amplitude and long propagation distance.Traditional satellite observations provide only twodimensional observations of ocean signatures induced by ISWs.The Surface Water and Ocean Topography(SWOT)satellite has drawn significant attention due to its high resolution and threedimensional observation capabilities.SWOT can generate high-precision three-dimensional sea surface topography,capture sea surface undulations,and reveal ISW-related surface oscillations,thus offering a new perspective for studying ISWs.We collected 43 SWOT observations with clear ISW signatures in the Lombok Strait from August 2023 to June 2024.Based on collected data,the ISW imaging characteristics and distributions were analyzed,and the ISW-related sea level anomaly(SLA)data were measured by the SWOT to calculate the ISW amplitude and reveal the amplitude variations during the propagation along the wave crest.The ISW amplitudes generally range between 10 and 100 m,with most ISW amplitudes between 20 and 40 m.By analyzing two consecutive generated ISW packets,we identified the spreading effect along ISW wave crests,which manifests as ISW amplitude decrease with increase in propagation distance,and the amplitude distribution is non-uniform along the wave crest.Further analysis of the propagation paths of the maximum amplitude of ISW moving northward through the Lombok Strait revealed that these maxima are predominantly oriented in northeast direction.Finally,the relationship between the amplitude of ISW and the resulting SLA was analyzed.The Pearson correlation coefficient between these two variables is as high as 0.90,which suggests a strong positive correlation between amplitude and SLA.Furthermore,this relationship is closely related to the water depth,indicating that the three-dimensional sea surface observations provided by SWOT offer crucial observational data for the inversion of amplitudes of ISW.
文摘ropic 1:Regarding sustainable development and global public interests,what should international Al standards focus on?James Ong:Since 2019,I have witnessed the evolution of WAIC and found that a consensus on the philosophical and ethic level on advocating“AI for humanity”is necessary,since ethics factor carries more weight in standards development.I want to emphasize three points:AI assisting sustainable development,AI empowering a balanced global development,and human-AI coordination for preventing AI risks.
基金supported by National Institute of Food and Agriculture,U.S.Department of Agriculture,under the award number 2024-67015-42622 to PFMississippi State Agricultural and Forestry Experiment Station(MAFES)Strategic Research Initiative Programsupported by the Mississippi State University College of Agriculture and Life Science/MAFES Undergraduate Research Scholars Program。
文摘Gut-brain communication via the peripheral neural network is vital for regulating local digestive function and systemic physiology.Gut microbiota,which produces a wide array of neuroactive compounds,is a critical modulator in this bidirectional dialog.Perturbations in the gut microbiota have been implicated in neurological disorders such as depression and stress.Distinct from humans and other monogastric animals,ruminants possess a unique,microbially dense gastrointestinal compartment,the rumen,that facilitates the digestion of fibrous plant materials.These ruminal microbes are likely key contributors to rumen-brain crosstalk.Unlike certain microbe-derived neuroactive compounds produced in the hindgut that are minimally absorbed and primarily excreted in feces,those generated in rumen can reach the small intestine,where they are largely absorbed and affect central nervous system through systemic regulation in addition to the vagal pathway.Notably,emerging evidence suggests that rumen microbiota dysbiosis under stress is associated with abnormal behavior,altered hormonal and neurotransmitter levels.In this review,we introduce the concept of the rumen-microbiome-brain axis by comparing the anatomical structures and microbial characteristics of the intestine and the rumen,emphasizing the neuroactive potential of rumen microbiome and underlying mechanisms.Advances in this frontier hold tremendous promise to reveal a novel dimension of the gut-microbiome-brain axis,providing transformative opportunities to improve ruminant welfare,productivity,and agricultural sustainability.
基金supported in part by NSFC under Grant 62422407in part by RGC under Grant 26204424in part by ACCESS–AI Chip Center for Emerging Smart Systems, sponsored by the Inno HK initiative of the Innovation and Technology Commission of the Hong Kong Special Administrative Region Government
文摘Robotic computing systems play an important role in enabling intelligent robotic tasks through intelligent algo-rithms and supporting hardware.In recent years,the evolution of robotic algorithms indicates a roadmap from traditional robotics to hierarchical and end-to-end models.This algorithmic advancement poses a critical challenge in achieving balanced system-wide performance.Therefore,algorithm-hardware co-design has emerged as the primary methodology,which ana-lyzes algorithm behaviors on hardware to identify common computational properties.These properties can motivate algo-rithm optimization to reduce computational complexity and hardware innovation from architecture to circuit for high performance and high energy efficiency.We then reviewed recent works on robotic and embodied AI algorithms and computing hard-ware to demonstrate this algorithm-hardware co-design methodology.In the end,we discuss future research opportunities by answering two questions:(1)how to adapt the computing platforms to the rapid evolution of embodied AI algorithms,and(2)how to transform the potential of emerging hardware innovations into end-to-end inference improvements.
基金funded by Ongoing Research Funding Program for Project number(ORF-2025-648),King Saud University,Riyadh,Saudi Arabia.
文摘Heart disease remains a leading cause of mortality worldwide,emphasizing the urgent need for reliable and interpretable predictive models to support early diagnosis and timely intervention.However,existing Deep Learning(DL)approaches often face several limitations,including inefficient feature extraction,class imbalance,suboptimal classification performance,and limited interpretability,which collectively hinder their deployment in clinical settings.To address these challenges,we propose a novel DL framework for heart disease prediction that integrates a comprehensive preprocessing pipeline with an advanced classification architecture.The preprocessing stage involves label encoding and feature scaling.To address the issue of class imbalance inherent in the personal key indicators of the heart disease dataset,the localized random affine shadowsampling technique is employed,which enhances minority class representation while minimizing overfitting.At the core of the framework lies the Deep Residual Network(DeepResNet),which employs hierarchical residual transformations to facilitate efficient feature extraction and capture complex,non-linear relationships in the data.Experimental results demonstrate that the proposed model significantly outperforms existing techniques,achieving improvements of 3.26%in accuracy,3.16%in area under the receiver operating characteristics,1.09%in recall,and 1.07%in F1-score.Furthermore,robustness is validated using 10-fold crossvalidation,confirming the model’s generalizability across diverse data distributions.Moreover,model interpretability is ensured through the integration of Shapley additive explanations and local interpretable model-agnostic explanations,offering valuable insights into the contribution of individual features to model predictions.Overall,the proposed DL framework presents a robust,interpretable,and clinically applicable solution for heart disease prediction.
基金funded by Multimedia University,Cyberjaya,Selangor,Malaysia(Grant Number:PostDoc(MMUI/240029)).
文摘Manual inspection of onba earing casting defects is not realistic and unreliable,particularly in the case of some micro-level anomalies which lead to major defects on a large scale.To address these challenges,we propose BearFusionNet,an attention-based deep learning architecture with multi-stream,which merges both DenseNet201 and MobileNetV2 for feature extraction with a classification head inspired by VGG19.This hybrid design,figuratively beaming from one layer to another,extracts the enormity of representations on different scales,backed by a prepreprocessing pipeline that brings defect saliency to the fore through contrast adjustment,denoising,and edge detection.The use of multi-head self-attention enhances feature fusion,enabling the model to capture both large and small spatial features.BearFusionNet achieves an accuracy of 99.66%and Cohen’s kappa score of 0.9929 in Kaggle’s Real-life Industrial Casting Defects dataset.Both McNemar’s and Wilcoxon signed-rank statistical tests,as well as fivefold cross-validation,are employed to assess the robustness of our proposed model.To interpret the model,we adopt Grad-Cam visualizations,which are the state of the art standard.Furthermore,we deploy BearFusionNet as a webbased system for near real-time inference(5-6 s per prediction),which enables the quickest yet accurate detection with visual explanations.Overall,BearFusionNet is an interpretable,accurate,and deployable solution that can automatically detect casting defects,leading to significant advances in the innovative industrial environment.
文摘It’s a pleasure to be here and speak about the industrial copilot and generative AI and the changing applications.As you all know,generative AI has arrived since a few years ago,we have generative AI not only in the consumer world,but also in the industrial world.Siemens is very active in the industrial space.We need to make AI real,because at the end in industry manufacturing,you are producing parts in the real world.So we need to make sure that AI and its applications can interact and comply with the real world.
基金supported by the Scientific Research Project of China Rehabilitation Research Center,No.2021zx-23the National Natural Science Foundation of China,No.32100925the Beijing Nova Program,No.Z211100002121038。
文摘Poststro ke cognitive impairment is a major secondary effect of ischemic stroke in many patients;however,few options are available for the early diagnosis and treatment of this condition.The aims of this study were to(1)determine the specific relationship between hypoxic andα-synuclein during the occur of poststroke cognitive impairment and(2)assess whether the serum phosphorylatedα-synuclein level can be used as a biomarker for poststro ke cognitive impairment.We found that the phosphorylatedα-synuclein level was significantly increased and showed pathological aggregation around the cerebral infa rct area in a mouse model of ischemic stroke.In addition,neuronalα-synuclein phosphorylation and aggregation were observed in the brain tissue of mice subjected to chronic hypoxia,suggesting that hypoxia is the underlying cause ofα-synuclein-mediated pathology in the brains of mice with ischemic stroke.Serum phosphorylatedα-synuclein levels in patients with ischemic stroke were significantly lower than those in healt hy subjects,and were positively correlated with cognition levels in patients with ischemic stroke.Furthermore,a decrease in serum high-density lipoprotein levels in stroke patie nts was significantly correlated with a decrease in phosphorylatedα-synuclein levels.Although ischemic stroke mice did not show significant cognitive impairment or disrupted lipid metabolism 14 days after injury,some of them exhibited decreased cognitive function and reduced phosphorylatedα-synuclein levels.Taken together,our results suggest that serum phosphorylatedα-synuclein is a potential biomarker for poststroke cognitive impairment.
基金Australian Research Council Project(FL-170100117).
文摘To avoid the laborious annotation process for dense prediction tasks like semantic segmentation,unsupervised domain adaptation(UDA)methods have been proposed to leverage the abundant annotations from a source domain,such as virtual world(e.g.,3D games),and adapt models to the target domain(the real world)by narrowing the domain discrepancies.However,because of the large domain gap,directly aligning two distinct domains without considering the intermediates leads to inefficient alignment and inferior adaptation.To address this issue,we propose a novel learnable evolutionary Category Intermediates(CIs)guided UDA model named Leci,which enables the information transfer between the two domains via two processes,i.e.,Distilling and Blending.Starting from a random initialization,the CIs learn shared category-wise semantics automatically from two domains in the Distilling process.Then,the learned semantics in the CIs are sent back to blend the domain features through a residual attentive fusion(RAF)module,such that the categorywise features of both domains shift towards each other.As the CIs progressively and consistently learn from the varying feature distributions during training,they are evolutionary to guide the model to achieve category-wise feature alignment.Experiments on both GTA5 and SYNTHIA datasets demonstrate Leci's superiority over prior representative methods.
