Deep learning(DL)has become a crucial technique for predicting the El Niño-Southern Oscillation(ENSO)and evaluating its predictability.While various DL-based models have been developed for ENSO predictions,many f...Deep learning(DL)has become a crucial technique for predicting the El Niño-Southern Oscillation(ENSO)and evaluating its predictability.While various DL-based models have been developed for ENSO predictions,many fail to capture the coherent multivariate evolution within the coupled ocean-atmosphere system of the tropical Pacific.To address this three-dimensional(3D)limitation and represent ENSO-related ocean-atmosphere interactions more accurately,a novel this 3D multivariate prediction model was proposed based on a Transformer architecture,which incorporates a spatiotemporal self-attention mechanism.This model,named 3D-Geoformer,offers several advantages,enabling accurate ENSO predictions up to one and a half years in advance.Furthermore,an integrated gradient method was introduced into the model to identify the sources of predictability for sea surface temperature(SST)variability in the eastern equatorial Pacific.Results reveal that the 3D-Geoformer effectively captures ENSO-related precursors during the evolution of ENSO events,particularly the thermocline feedback processes and ocean temperature anomaly pathways on and off the equator.By extending DL-based ENSO predictions from one-dimensional Niño time series to 3D multivariate fields,the 3D-Geoformer represents a significant advancement in ENSO prediction.This study provides details in the model formulation,analysis procedures,sensitivity experiments,and illustrative examples,offering practical guidance for the application of the model in ENSO research.展开更多
The osteochondral defect repair has been most extensively studied due to the rising demand for new therapies to diseases such as osteoarthritis.Tissue engineering has been proposed as a promising strategy to meet the ...The osteochondral defect repair has been most extensively studied due to the rising demand for new therapies to diseases such as osteoarthritis.Tissue engineering has been proposed as a promising strategy to meet the demand of simultaneous regeneration of both cartilage and subchondral bone by constructing integrated gradient tissue-engineered osteochondral scaffold(IGTEOS).This review brought forward the main challenges of establishing a satisfactory IGTEOS from the perspectives of the complexity of physiology and microenvironment of osteochondral tissue,and the limitations of obtaining the desired and required scaffold.Then,we comprehensively discussed and summarized the current tissue-engineered efforts to resolve the above challenges,including architecture strategies,fabrication techniques and in vitro/in vivo evaluation methods of the IGTEOS.Especially,we highlighted the advantages and limitations of various fabrication techniques of IGTEOS,and common cases of IGTEOS application.Finally,based on the above challenges and current research progress,we analyzed in details the future perspectives of tissue-engineered osteochondral construct,so as to achieve the perfect reconstruction of the cartilaginous and osseous layers of osteochondral tissue simultaneously.This comprehensive and instructive review could provide deep insights into our current understanding of IGTEOS.展开更多
This paper extends the one-dimensional(1D)nonlocal strain gradient integral model(NStraGIM)to the two-dimensional(2D)Kirchhoff axisymmetric nanoplates,based on nonlocal strain gradient integral relations formulated al...This paper extends the one-dimensional(1D)nonlocal strain gradient integral model(NStraGIM)to the two-dimensional(2D)Kirchhoff axisymmetric nanoplates,based on nonlocal strain gradient integral relations formulated along both the radial and circumferential directions.By transforming the proposed integral constitutive equations into the equivalent differential forms,complemented by the corresponding constitutive boundary conditions(CBCs),a well-posed mathematical formulation is established for analyzing the axisymmetric bending and buckling of annular/circular functionally graded(FG)sandwich nanoplates.The boundary conditions at the inner edge of a solid nanoplate are derived by L'H?spital's rule.The numerical solution is obtained by the generalized differential quadrature method(GDQM).The accuracy of the proposed model is validated through comparison with the data from the existing literature.A parameter study is conducted to demonstrate the effects of FG sandwich parameters,size parameters,and nonlocal gradient parameters.展开更多
We present a study on the dynamic stability of porous functionally graded(PFG)beams under hygro-thermal loading.The variations of the properties of the beams across the beam thicknesses are described by the power-law ...We present a study on the dynamic stability of porous functionally graded(PFG)beams under hygro-thermal loading.The variations of the properties of the beams across the beam thicknesses are described by the power-law model.Unlike most studies on this topic,we consider both the bending deformation of the beams and the hygro-thermal load as size-dependent,simultaneously,by adopting the equivalent differential forms of the well-posed nonlocal strain gradient integral theory(NSGIT)which are strictly equipped with a set of constitutive boundary conditions(CBCs),and through which both the stiffness-hardening and stiffness-softening effects of the structures can be observed with the length-scale parameters changed.All the variables presented in the differential problem formulation are discretized.The numerical solution of the dynamic instability region(DIR)of various bounded beams is then developed via the generalized differential quadrature method(GDQM).After verifying the present formulation and results,we examine the effects of different parameters such as the nonlocal/gradient length-scale parameters,the static force factor,the functionally graded(FG)parameter,and the porosity parameter on the DIR.Furthermore,the influence of considering the size-dependent hygro-thermal load is also presented.展开更多
An integral nonlocal stress gradient viscoelastic model is proposed on the basis of the integral nonlocal stress gradient model and the standard viscoelastic model,and is utilized to investigate the free damping vibra...An integral nonlocal stress gradient viscoelastic model is proposed on the basis of the integral nonlocal stress gradient model and the standard viscoelastic model,and is utilized to investigate the free damping vibration analysis of the viscoelastic BernoulliEuler microbeams in thermal environment.Hamilton's principle is used to derive the differential governing equations and corresponding boundary conditions.The integral relations between the strain and the nonlocal stress are converted into a differential form with constitutive constraints.The size-dependent axial thermal stress due to the variation of the environmental temperature is derived explicitly.The Laplace transformation is utilized to obtain the explicit expression for the bending deflection and moment.Considering the boundary conditions and constitutive constraints,one can get a nonlinear equation with complex coefficients,from which the complex characteristic frequency can be determined.A two-step numerical method is proposed to solve the elastic vibration frequency and the damping ratio.The effects of length scale parameters,viscous coefficient,thermal stress,vibration order on the vibration frequencies,and critical viscous coefficient are investigated numerically for the viscoelastic Bernoulli-Euler microbeams under different boundary conditions.展开更多
Negative logarithm of the acid dissociation constant(pK_(a))significantly influences the absorption,dis-tribution,metabolism,excretion,and toxicity(ADMET)properties of molecules and is a crucial indicator in drug rese...Negative logarithm of the acid dissociation constant(pK_(a))significantly influences the absorption,dis-tribution,metabolism,excretion,and toxicity(ADMET)properties of molecules and is a crucial indicator in drug research.Given the rapid and accurate characteristics of computational methods,their role in predicting drug properties is increasingly important.Although many pK_(a) prediction models currently exist,they often focus on enhancing model precision while neglecting interpretability.In this study,we present GraFpKa,a pK_(a) prediction model using graph neural networks(GNNs)and molecular finger-prints.The results show that our acidic and basic models achieved mean absolute errors(MAEs)of 0.621 and 0.402,respectively,on the test set,demonstrating good predictive performance.Notably,to improve interpretability,GraFpKa also incorporates Integrated Gradients(IGs),providing a clearer visual description of the atoms significantly affecting the pK_(a) values.The high reliability and interpretability of GraFpKa ensure accurate pKa predictions while also facilitating a deeper understanding of the relation-ship between molecular structure and pK_(a) values,making it a valuable tool in the field of pK_(a) prediction.展开更多
Background:Convolutional neural networks(CNN)have achieved remarkable success in medical image analysis.However,unlike some general-domain tasks where model accuracy is paramount,medical applications demand both accur...Background:Convolutional neural networks(CNN)have achieved remarkable success in medical image analysis.However,unlike some general-domain tasks where model accuracy is paramount,medical applications demand both accuracy and explainability due to the high stakes affecting patients'lives.Based on model explanations,clinicians can evaluate the diagnostic decisions suggested by CNN.Nevertheless,prior explainable artificial intelligence methods treat medical image tasks akin to general vision tasks,following end-to-end paradigms to generate explanations and frequently overlooking crucial clinical domain knowledge.Methods:We propose a plug-and-play module that explicitly integrates anatomic boundary information into the explanation process for CNN-based thoracopathy classifiers.To generate the anatomic boundary of the lung parenchyma,we utilize a lung segmentation model developed on external public datasets and deploy it on the unseen target dataset to constrain model ex-planations within the lung parenchyma for the clinical task of thoracopathy classification.Results:Assessed by the intersection over union and dice similarity coefficient between model-extracted explanations and expert-annotated lesion areas,our method consistently outperformed the baseline devoid of clinical domain knowledge in 71 out of 72 scenarios,encompassing 3 CNN architectures(VGG-11,ResNet-18,and AlexNet),2 classification settings(binary and multi-label),3 explanation methods(Saliency Map,Grad-CAM,and Integrated Gradients),and 4 co-occurred thoracic diseases(Atelectasis,Fracture,Mass,and Pneumothorax).Conclusions:We underscore the effectiveness of leveraging radiology knowledge in improving model explanations for CNN and envisage that it could inspire future efforts to integrate clinical domain knowledge into medical image analysis.展开更多
Roof falls due to geological conditions are major hazards in the mining industry,causing work time loss,injuries,and fatalities.There are roof fall problems caused by high horizontal stress in several largeopening lim...Roof falls due to geological conditions are major hazards in the mining industry,causing work time loss,injuries,and fatalities.There are roof fall problems caused by high horizontal stress in several largeopening limestone mines in the eastern and midwestern United States.The typical hazard management approach for this type of roof fall hazards relies heavily on visual inspections and expert knowledge.In this context,we proposed a deep learning system for detection of the roof fall hazards caused by high horizontal stress.We used images depicting hazardous and non-hazardous roof conditions to develop a convolutional neural network(CNN)for autonomous detection of hazardous roof conditions.To compensate for limited input data,we utilized a transfer learning approach.In the transfer learning approach,an already-trained network is used as a starting point for classification in a similar domain.Results show that this approach works well for classifying roof conditions as hazardous or safe,achieving a statistical accuracy of 86.4%.This result is also compared with a random forest classifier,and the deep learning approach is more successful at classification of roof conditions.However,accuracy alone is not enough to ensure a reliable hazard management system.System constraints and reliability are improved when the features used by the network are understood.Therefore,we used a deep learning interpretation technique called integrated gradients to identify the important geological features in each image for prediction.The analysis of integrated gradients shows that the system uses the same roof features as the experts do on roof fall hazards detection.The system developed in this paper demonstrates the potential of deep learning in geotechnical hazard management to complement human experts,and likely to become an essential part of autonomous operations in cases where hazard identification heavily depends on expert knowledge.Moreover,deep learning-based systems reduce expert exposure to hazardous conditions.展开更多
We consider solving integral equations of the second kind defined on the half-line [0, infinity) by the preconditioned conjugate gradient method. Convergence is known to be slow due to the non-compactness of the assoc...We consider solving integral equations of the second kind defined on the half-line [0, infinity) by the preconditioned conjugate gradient method. Convergence is known to be slow due to the non-compactness of the associated integral operator. In this paper, we construct two different circulant integral operators to be used as preconditioners for the method to speed up its convergence rate. We prove that if the given integral operator is close to a convolution-type integral operator, then the preconditioned systems will have spectrum clustered around 1 and hence the preconditioned conjugate gradient method will converge superlinearly. Numerical examples are given to illustrate the fast convergence.展开更多
In this paper,based on discrete gradient,a dissipation-preserving integrator for weakly dissipative perturbations of oscillatory Hamiltonian system is established.The solution of this system is a damped nonlinear osci...In this paper,based on discrete gradient,a dissipation-preserving integrator for weakly dissipative perturbations of oscillatory Hamiltonian system is established.The solution of this system is a damped nonlinear oscillator.Basically,lots of nonlinear oscillatory mechanical systems including frictional forces lend themselves to this approach.The new integrator gives a discrete analogue of the dissipation property of the original system.Meanwhile,since the integrator is based on the variation-of-constants formula for oscillatory systems,it preserves the oscillatory structure of the system.Some properties of the new integrator are derived.The convergence is analyzed for the implicit iterations based on the discrete gradient integrator,and it turns out that the convergence of the implicit iterations based on the new integrator is independent of k Mk,where M governs the main oscillation of the system and usually k Mk≫1.This significant property shows that a larger stepsize can be chosen for the new schemes than that for the traditional discrete gradient integrators when applied to the oscillatory Hamiltonian system.Numerical experiments are carried out to show the effectiveness and efficiency of the new integrator in comparison with the traditional discrete gradient methods in the scientific literature。展开更多
基金Supported by the Laoshan Laboratory(No.LSKJ202202402)the National Natural Science Foundation of China(No.42030410)+2 种基金the Startup Foundation for Introducing Talent of Nanjing University of Information Science&Technology,and Jiangsu Innovation Research Group(No.JSSCTD 202346)supported by the China National Postdoctoral Program for Innovative Talents(No.BX20240169)the China Postdoctoral Science Foundation(No.2141062400101)。
文摘Deep learning(DL)has become a crucial technique for predicting the El Niño-Southern Oscillation(ENSO)and evaluating its predictability.While various DL-based models have been developed for ENSO predictions,many fail to capture the coherent multivariate evolution within the coupled ocean-atmosphere system of the tropical Pacific.To address this three-dimensional(3D)limitation and represent ENSO-related ocean-atmosphere interactions more accurately,a novel this 3D multivariate prediction model was proposed based on a Transformer architecture,which incorporates a spatiotemporal self-attention mechanism.This model,named 3D-Geoformer,offers several advantages,enabling accurate ENSO predictions up to one and a half years in advance.Furthermore,an integrated gradient method was introduced into the model to identify the sources of predictability for sea surface temperature(SST)variability in the eastern equatorial Pacific.Results reveal that the 3D-Geoformer effectively captures ENSO-related precursors during the evolution of ENSO events,particularly the thermocline feedback processes and ocean temperature anomaly pathways on and off the equator.By extending DL-based ENSO predictions from one-dimensional Niño time series to 3D multivariate fields,the 3D-Geoformer represents a significant advancement in ENSO prediction.This study provides details in the model formulation,analysis procedures,sensitivity experiments,and illustrative examples,offering practical guidance for the application of the model in ENSO research.
基金support from the National Natural Science Foundation of China(No.32171345)Hebei Provincial Natural Science Foundation of China(No.C2022104003)+2 种基金the Fok Ying Tung Education Foundation(No.141039)the Fund of Key Laboratory of Advanced Materials of Ministry of Education,the International Joint Research Center of Aerospace Biotechnology and Medical Engineering,Ministry of Science and Technology of Chinathe 111 Project(No.B13003).
文摘The osteochondral defect repair has been most extensively studied due to the rising demand for new therapies to diseases such as osteoarthritis.Tissue engineering has been proposed as a promising strategy to meet the demand of simultaneous regeneration of both cartilage and subchondral bone by constructing integrated gradient tissue-engineered osteochondral scaffold(IGTEOS).This review brought forward the main challenges of establishing a satisfactory IGTEOS from the perspectives of the complexity of physiology and microenvironment of osteochondral tissue,and the limitations of obtaining the desired and required scaffold.Then,we comprehensively discussed and summarized the current tissue-engineered efforts to resolve the above challenges,including architecture strategies,fabrication techniques and in vitro/in vivo evaluation methods of the IGTEOS.Especially,we highlighted the advantages and limitations of various fabrication techniques of IGTEOS,and common cases of IGTEOS application.Finally,based on the above challenges and current research progress,we analyzed in details the future perspectives of tissue-engineered osteochondral construct,so as to achieve the perfect reconstruction of the cartilaginous and osseous layers of osteochondral tissue simultaneously.This comprehensive and instructive review could provide deep insights into our current understanding of IGTEOS.
基金Project supported by the National Natural Science Foundation of China(No.12172169)the Priority Academic Program Development of Jiangsu Higher Education Institutions。
文摘This paper extends the one-dimensional(1D)nonlocal strain gradient integral model(NStraGIM)to the two-dimensional(2D)Kirchhoff axisymmetric nanoplates,based on nonlocal strain gradient integral relations formulated along both the radial and circumferential directions.By transforming the proposed integral constitutive equations into the equivalent differential forms,complemented by the corresponding constitutive boundary conditions(CBCs),a well-posed mathematical formulation is established for analyzing the axisymmetric bending and buckling of annular/circular functionally graded(FG)sandwich nanoplates.The boundary conditions at the inner edge of a solid nanoplate are derived by L'H?spital's rule.The numerical solution is obtained by the generalized differential quadrature method(GDQM).The accuracy of the proposed model is validated through comparison with the data from the existing literature.A parameter study is conducted to demonstrate the effects of FG sandwich parameters,size parameters,and nonlocal gradient parameters.
基金Project supported by the National Natural Science Foundation of China(No.12172169)the Natural Sciences and Engineering Research Council of Canada(No.NSERC RGPIN-2023-03227)。
文摘We present a study on the dynamic stability of porous functionally graded(PFG)beams under hygro-thermal loading.The variations of the properties of the beams across the beam thicknesses are described by the power-law model.Unlike most studies on this topic,we consider both the bending deformation of the beams and the hygro-thermal load as size-dependent,simultaneously,by adopting the equivalent differential forms of the well-posed nonlocal strain gradient integral theory(NSGIT)which are strictly equipped with a set of constitutive boundary conditions(CBCs),and through which both the stiffness-hardening and stiffness-softening effects of the structures can be observed with the length-scale parameters changed.All the variables presented in the differential problem formulation are discretized.The numerical solution of the dynamic instability region(DIR)of various bounded beams is then developed via the generalized differential quadrature method(GDQM).After verifying the present formulation and results,we examine the effects of different parameters such as the nonlocal/gradient length-scale parameters,the static force factor,the functionally graded(FG)parameter,and the porosity parameter on the DIR.Furthermore,the influence of considering the size-dependent hygro-thermal load is also presented.
基金Project supported by the National Natural Science Foundation of China(No.12172169)。
文摘An integral nonlocal stress gradient viscoelastic model is proposed on the basis of the integral nonlocal stress gradient model and the standard viscoelastic model,and is utilized to investigate the free damping vibration analysis of the viscoelastic BernoulliEuler microbeams in thermal environment.Hamilton's principle is used to derive the differential governing equations and corresponding boundary conditions.The integral relations between the strain and the nonlocal stress are converted into a differential form with constitutive constraints.The size-dependent axial thermal stress due to the variation of the environmental temperature is derived explicitly.The Laplace transformation is utilized to obtain the explicit expression for the bending deflection and moment.Considering the boundary conditions and constitutive constraints,one can get a nonlinear equation with complex coefficients,from which the complex characteristic frequency can be determined.A two-step numerical method is proposed to solve the elastic vibration frequency and the damping ratio.The effects of length scale parameters,viscous coefficient,thermal stress,vibration order on the vibration frequencies,and critical viscous coefficient are investigated numerically for the viscoelastic Bernoulli-Euler microbeams under different boundary conditions.
基金upported by the National Key Research and Development Program of China(Grant No.:2023YFF1204904)the National Natural Science Foundation of China(Grant Nos.:U23A20530 and 82173746)Shanghai Frontiers Science Center of Optogenetic Techniques for Cell Metabolism(Shanghai Municipal Education Commission,China).
文摘Negative logarithm of the acid dissociation constant(pK_(a))significantly influences the absorption,dis-tribution,metabolism,excretion,and toxicity(ADMET)properties of molecules and is a crucial indicator in drug research.Given the rapid and accurate characteristics of computational methods,their role in predicting drug properties is increasingly important.Although many pK_(a) prediction models currently exist,they often focus on enhancing model precision while neglecting interpretability.In this study,we present GraFpKa,a pK_(a) prediction model using graph neural networks(GNNs)and molecular finger-prints.The results show that our acidic and basic models achieved mean absolute errors(MAEs)of 0.621 and 0.402,respectively,on the test set,demonstrating good predictive performance.Notably,to improve interpretability,GraFpKa also incorporates Integrated Gradients(IGs),providing a clearer visual description of the atoms significantly affecting the pK_(a) values.The high reliability and interpretability of GraFpKa ensure accurate pKa predictions while also facilitating a deeper understanding of the relation-ship between molecular structure and pK_(a) values,making it a valuable tool in the field of pK_(a) prediction.
文摘Background:Convolutional neural networks(CNN)have achieved remarkable success in medical image analysis.However,unlike some general-domain tasks where model accuracy is paramount,medical applications demand both accuracy and explainability due to the high stakes affecting patients'lives.Based on model explanations,clinicians can evaluate the diagnostic decisions suggested by CNN.Nevertheless,prior explainable artificial intelligence methods treat medical image tasks akin to general vision tasks,following end-to-end paradigms to generate explanations and frequently overlooking crucial clinical domain knowledge.Methods:We propose a plug-and-play module that explicitly integrates anatomic boundary information into the explanation process for CNN-based thoracopathy classifiers.To generate the anatomic boundary of the lung parenchyma,we utilize a lung segmentation model developed on external public datasets and deploy it on the unseen target dataset to constrain model ex-planations within the lung parenchyma for the clinical task of thoracopathy classification.Results:Assessed by the intersection over union and dice similarity coefficient between model-extracted explanations and expert-annotated lesion areas,our method consistently outperformed the baseline devoid of clinical domain knowledge in 71 out of 72 scenarios,encompassing 3 CNN architectures(VGG-11,ResNet-18,and AlexNet),2 classification settings(binary and multi-label),3 explanation methods(Saliency Map,Grad-CAM,and Integrated Gradients),and 4 co-occurred thoracic diseases(Atelectasis,Fracture,Mass,and Pneumothorax).Conclusions:We underscore the effectiveness of leveraging radiology knowledge in improving model explanations for CNN and envisage that it could inspire future efforts to integrate clinical domain knowledge into medical image analysis.
基金partially supported by the National Institute for Occupational Safety and Health,contract number 0000HCCR-2019-36403。
文摘Roof falls due to geological conditions are major hazards in the mining industry,causing work time loss,injuries,and fatalities.There are roof fall problems caused by high horizontal stress in several largeopening limestone mines in the eastern and midwestern United States.The typical hazard management approach for this type of roof fall hazards relies heavily on visual inspections and expert knowledge.In this context,we proposed a deep learning system for detection of the roof fall hazards caused by high horizontal stress.We used images depicting hazardous and non-hazardous roof conditions to develop a convolutional neural network(CNN)for autonomous detection of hazardous roof conditions.To compensate for limited input data,we utilized a transfer learning approach.In the transfer learning approach,an already-trained network is used as a starting point for classification in a similar domain.Results show that this approach works well for classifying roof conditions as hazardous or safe,achieving a statistical accuracy of 86.4%.This result is also compared with a random forest classifier,and the deep learning approach is more successful at classification of roof conditions.However,accuracy alone is not enough to ensure a reliable hazard management system.System constraints and reliability are improved when the features used by the network are understood.Therefore,we used a deep learning interpretation technique called integrated gradients to identify the important geological features in each image for prediction.The analysis of integrated gradients shows that the system uses the same roof features as the experts do on roof fall hazards detection.The system developed in this paper demonstrates the potential of deep learning in geotechnical hazard management to complement human experts,and likely to become an essential part of autonomous operations in cases where hazard identification heavily depends on expert knowledge.Moreover,deep learning-based systems reduce expert exposure to hazardous conditions.
文摘We consider solving integral equations of the second kind defined on the half-line [0, infinity) by the preconditioned conjugate gradient method. Convergence is known to be slow due to the non-compactness of the associated integral operator. In this paper, we construct two different circulant integral operators to be used as preconditioners for the method to speed up its convergence rate. We prove that if the given integral operator is close to a convolution-type integral operator, then the preconditioned systems will have spectrum clustered around 1 and hence the preconditioned conjugate gradient method will converge superlinearly. Numerical examples are given to illustrate the fast convergence.
基金supported in part by the Natural Science Foundation of China under Grant 11701271.
文摘In this paper,based on discrete gradient,a dissipation-preserving integrator for weakly dissipative perturbations of oscillatory Hamiltonian system is established.The solution of this system is a damped nonlinear oscillator.Basically,lots of nonlinear oscillatory mechanical systems including frictional forces lend themselves to this approach.The new integrator gives a discrete analogue of the dissipation property of the original system.Meanwhile,since the integrator is based on the variation-of-constants formula for oscillatory systems,it preserves the oscillatory structure of the system.Some properties of the new integrator are derived.The convergence is analyzed for the implicit iterations based on the discrete gradient integrator,and it turns out that the convergence of the implicit iterations based on the new integrator is independent of k Mk,where M governs the main oscillation of the system and usually k Mk≫1.This significant property shows that a larger stepsize can be chosen for the new schemes than that for the traditional discrete gradient integrators when applied to the oscillatory Hamiltonian system.Numerical experiments are carried out to show the effectiveness and efficiency of the new integrator in comparison with the traditional discrete gradient methods in the scientific literature。
