arXiv subjects

Feng Xie

Publications and source records attributed to Feng Xie.

At least 73 records · Page 4Linked to original sources

arXiv2021

Uniform regularity estimates and invisicid limit for the compressible non-resistive magnetohydrodynamics system

We are concerned with the uniform regularity estimates of solutions to the two dimensional compressible non-resistive magnetohydrodynamics (MHD) equations with the no-slip boundary condition on velocity in the half plane. Under the assumption that the initial magnetic field is transverse to the boundary, the uniform conormal energy estimates are established for the solutions to compressible MHD equations with respect to small viscosity coefficients. As a direct consequence, we proved the inviscid limit of solutions from viscous MHD systems to the ideal MHD systems in $L^\infty$ sense. It shows that the transverse magnetic field can prevent the boundary layers from occurring in some physical regime.

math.AP↗

arXiv2021

Deep learning for temporal data representation in electronic health records: A systematic review of challenges and methodologies

Objective: Temporal electronic health records (EHRs) can be a wealth of information for secondary uses, such as clinical events prediction or chronic disease management. However, challenges exist for temporal data representation. We therefore sought to identify these challenges and evaluate novel methodologies for addressing them through a systematic examination of deep learning solutions. Methods: We searched five databases (PubMed, EMBASE, the Institute of Electrical and Electronics Engineers [IEEE] Xplore Digital Library, the Association for Computing Machinery [ACM] digital library, and Web of Science) complemented with hand-searching in several prestigious computer science conference proceedings. We sought articles that reported deep learning methodologies on temporal data representation in structured EHR data from January 1, 2010, to August 30, 2020. We summarized and analyzed the selected articles from three perspectives: nature of time series, methodology, and model implementation. Results: We included 98 articles related to temporal data representation using deep learning. Four major challenges were identified, including data irregularity, data heterogeneity, data sparsity, and model opacity. We then studied how deep learning techniques were applied to address these challenges. Finally, we discuss some open challenges arising from deep learning. Conclusion: Temporal EHR data present several major challenges for clinical prediction modeling and data utilization. To some extent, current deep learning solutions can address these challenges. Future studies can consider designing comprehensive and integrated solutions. Moreover, researchers should incorporate additional clinical domain knowledge into study designs and enhance the interpretability of the model to facilitate its implementation in clinical practice.

cs.LG↗

arXiv2021

AutoScore-Imbalance: An interpretable machine learning tool for development of clinical scores with rare events data

Background: Medical decision-making impacts both individual and public health. Clinical scores are commonly used among a wide variety of decision-making models for determining the degree of disease deterioration at the bedside. AutoScore was proposed as a useful clinical score generator based on machine learning and a generalized linear model. Its current framework, however, still leaves room for improvement when addressing unbalanced data of rare events. Methods: Using machine intelligence approaches, we developed AutoScore-Imbalance, which comprises three components: training dataset optimization, sample weight optimization, and adjusted AutoScore. All scoring models were evaluated on the basis of their area under the curve (AUC) in the receiver operating characteristic analysis and balanced accuracy (i.e., mean value of sensitivity and specificity). By utilizing a publicly accessible dataset from Beth Israel Deaconess Medical Center, we assessed the proposed model and baseline approaches in the prediction of inpatient mortality. Results: AutoScore-Imbalance outperformed baselines in terms of AUC and balanced accuracy. The nine-variable AutoScore-Imbalance sub-model achieved the highest AUC of 0.786 (0.732-0.839) while the eleven-variable original AutoScore obtained an AUC of 0.723 (0.663-0.783), and the logistic regression with 21 variables obtained an AUC of 0.743 (0.685-0.800). The AutoScore-Imbalance sub-model (using down-sampling algorithm) yielded an AUC of 0. 0.771 (0.718-0.823) with only five variables, demonstrating a good balance between performance and variable sparsity. Conclusions: The AutoScore-Imbalance tool has the potential to be applied to highly unbalanced datasets to gain further insight into rare medical events and to facilitate real-world clinical decision-making.

cs.LG↗

arXiv2021

Inviscid Limit of Compressible Viscoelastic Equations with the No-Slip Boundary Condition

The inviscid limit for the two-dimensional compressible viscoelastic equations on the half plane is considered under the no-slip boundary condition. When the initial deformation tensor is a perturbation of the identity matrix and the initial density is near a positive constant, we establish the uniform estimates of solutions to the compressible viscoelastic flows in the conormal Sobolev spaces. It is well-known that for the corresponding inviscid limit of the compressible Navier-Stokes equations with the no-slip boundary condition, one does not expect the uniform energy estimates of solutions due to the appearance of strong boundary layers. However, when the deformation tensor effect is taken into account, our results show that the deformation tensor plays an important role in the vanishing viscosity process and can prevent the formation of strong boundary layers. As a result we are able to justify the inviscid limit of solutions for the compressible viscous flows under the no-slip boundary condition governed by the viscoelastic equations, based on the uniform conormal regularity estimates achieved in this paper.

math.AP↗

arXiv2021

AutoScore-Survival: Developing interpretable machine learning-based time-to-event scores with right-censored survival data

Scoring systems are highly interpretable and widely used to evaluate time-to-event outcomes in healthcare research. However, existing time-to-event scores are predominantly created ad-hoc using a few manually selected variables based on clinician's knowledge, suggesting an unmet need for a robust and efficient generic score-generating method. AutoScore was previously developed as an interpretable machine learning score generator, integrated both machine learning and point-based scores in the strong discriminability and accessibility. We have further extended it to time-to-event data and developed AutoScore-Survival, for automatically generating time-to-event scores with right-censored survival data. Random survival forest provides an efficient solution for selecting variables, and Cox regression was used for score weighting. We illustrated our method in a real-life study of 90-day mortality of patients in intensive care units and compared its performance with survival models (i.e., Cox) and the random survival forest. The AutoScore-Survival-derived scoring model was more parsimonious than survival models built using traditional variable selection methods (e.g., penalized likelihood approach and stepwise variable selection), and its performance was comparable to survival models using the same set of variables. Although AutoScore-Survival achieved a comparable integrated area under the curve of 0.782 (95% CI: 0.767-0.794), the integer-valued time-to-event scores generated are favorable in clinical applications because they are easier to compute and interpret. Our proposed AutoScore-Survival provides an automated, robust and easy-to-use machine learning-based clinical score generator to studies of time-to-event outcomes. It provides a systematic guideline to facilitate the future development of time-to-event scores for clinical applications.

cs.LG↗

arXiv2021

Global Solvability of 2D MHD Boundary Layer Equations in Analytic Function Spaces

In this paper we are concerned with the global well-posedness of solutions to magnetohydrodynamics (MHD) boundary layer equations in analytic function spaces. When the initial data is a small perturbation around a selected profile, and such a profile is governed by an one dimensional heat equation with a source term, we establish the global in time existence and uniqueness of analytic solutions to the two dimensional (2D) MHD boundary layer equations. It is noted that the far-field state of velocity is not required to be small initially, but decays to zero as time tends to infinity with suitable decay rates. The whole analysis is divided into two parts: When the initial far-field states are not small, we reformulate the original problem into a small perturbation problem by extracting a suitable background profile which is governed by a heat equation, then we prove a long-time existence of solutions, and the lower bound of lifespan of solutions is given in term of the small parameter of initial perturbation; After establishing the long-time existence, combining the decay properties of far-field state and the properties of solutions to the heat equation, it is shown that both the solution obtained in the first part and the background profile indeed satisfy some smallness requirements when time goes to the lower bound of lifespan of solutions. Then we can show the global existence of solutions after this moment.

math.AP↗

arXiv2020

Generalized Independent Noise Condition for Estimating Latent Variable Causal Graphs

Causal discovery aims to recover causal structures or models underlying the observed data. Despite its success in certain domains, most existing methods focus on causal relations between observed variables, while in many scenarios the observed ones may not be the underlying causal variables (e.g., image pixels), but are generated by latent causal variables or confounders that are causally related. To this end, in this paper, we consider Linear, Non-Gaussian Latent variable Models (LiNGLaMs), in which latent confounders are also causally related, and propose a Generalized Independent Noise (GIN) condition to estimate such latent variable graphs. Specifically, for two observed random vectors $\mathbf{Y}$ and $\mathbf{Z}$, GIN holds if and only if $ω^{\intercal}\mathbf{Y}$ and $\mathbf{Z}$ are statistically independent, where $ω$ is a parameter vector characterized from the cross-covariance between $\mathbf{Y}$ and $\mathbf{Z}$. From the graphical view, roughly speaking, GIN implies that causally earlier latent common causes of variables in $\mathbf{Y}$ d-separate $\mathbf{Y}$ from $\mathbf{Z}$. Interestingly, we find that the independent noise condition, i.e., if there is no confounder, causes are independent from the error of regressing the effect on the causes, can be seen as a special case of GIN. Moreover, we show that GIN helps locate latent variables and identify their causal structure, including causal directions. We further develop a recursive learning algorithm to achieve these goals. Experimental results on synthetic and real-world data demonstrate the effectiveness of our method.

cs.LG↗

arXiv2020

Electromagnetic Wave Propagation in GLHUA Invisible Sphere by GL No Scattering Full Wave Modeling and Inversion

Using GL no scattering full wave modeling and inversion, we create a GLHUA pre cloak electromagnetic (EM) material in the virtual sphere that makes the sphere is invisible. The invisible sphere is called GLHUA sphere. In GLHUA sphere, the Pre cloak relative parameter is not less than 1; the parameters and their derivative are continuous across the boundary r=R2 and the parameters are going to infinity at origin r=0. The phase velocity of EM wave in the sphere is less than light speed and going to zero at origin. The EM wave field excited in the outside of the sphere can not be disturbed by GLHUA sphere. By GL full wave method, we rigorously proved the incident EM wave field excited in outside of GLHUA sphere and propagation through the sphere without any scattering by the sphere, the total EM field in outside of the sphere equal to the incident wave field. Moreover, we prove that in GLHUA sphere with the pre cloak material, when r is going to origin, EM wave field propagation in GLHUA sphere is going to zero. We propose a $N$ dimensional MAXWELL equations. All copyright and patent of the GLHUA EM cloaks,GLHUA sphere and GL modeling and inversion methods are reserved by authors in GL Geophysical Laboratory.

physics.class-ph↗

arXiv2020

Practicable GLHUA Invisible Cloaks Absorb the Incident Wave and create Outgoing Wave Without Exceeding Light Speed Propagation and New N Dimensional Maxwell Equations

In this paper, we propose practicable GLHUA invisible cloaks, our cloak absorb the incident wave and create outgoing wave that make them to be invisible cloaks and analytical electromagnetic (EM) wave without exceeding light speed propagation. Discoveries and creations are reported. 1. New GLHUA-i,i=1,2,3 annular layer EM invisible cloak with relative refractive index large or equal to 1. 2. Analytical EM wave solution of EM wave equation with the GLHUA 1-3 cloak materials are found, In GLHUA-1 cloak, $\varepsilon_r = μ_r = \frac{R_2 ^2 }{r^2 }$, $\varepsilon_θ= \varepsilon_ϕ= μ_θ= μ_ϕ= \frac{R_2 - R_1 }{r - R_1 }$, In GLHUA-2 cloak, $\varepsilon _r = μ_r = \frac{1}{r^2 }\frac{(R_1 (r - R_1 ) + (R_2 - R_1 )^2 )^2 }{(R_2 - R_1 )^2 }$,. 3. A GLHUA expansion method and an exact analytical EM wave propagation in the GLHUA-i,i=1,2,3 cloak without exceeding light speed propagation that are propsed. 4.Novel negative space is proposed. 5. GLHUANP-i, i=2,3 transformation are proposed, which maps $ - \infty $ to $R_1$ and $-R_2$ to $R_2$. 6. The GLHUANP-i, i=2,3 transformation creates GLHUA-2 and GLHUA-3 cloaks. 7. GLHUA-i, i=1,2,3 cloaks absorb the incident wave and create outgoing wave that make them invisible cloak and analytical EM wave through them. 8. GLHUAF transformation and .GLHUAF invisible cloak with double negative materials is created. 9. The relative parameters of the GLHUA-1 are large than 1, radial parameter in GLHUA-2 cloak is large than positive number, refractive index of the GLHUA 1-3 cloak are large than 1. 10. Two wave fronts in GLHUA cloaks one front is absorbing incoming incident, other front is created wave by cloak materials. 11. New ND Maxwell Eq. is created. The patent, copyright and all rights are reserved by authors in GLGEO in USA.

physics.gen-ph↗

arXiv2020

Magnetic effects on the solvability of 2D MHD boundary layer equations without resistivity in Sobolev spaces

In this paper, we are concerned with the magnetic effect on the Sobolev solvability of boundary layer equations for the 2D incompressible MHD system without resistivity. The MHD boundary layer is described by the Prandtl type equations derived from the incompressible viscous MHD system without resistivity under the no-slip boundary condition on the velocity. Assuming that the initial tangential magnetic field does not degenerate, a local-in-time well-posedness in Sobolev spaces is proved without the monotonicity condition on the velocity field. Moreover, we show that if the tangential magnetic field shear layer is degenerate at one point, then the linearized MHD boundary layer system around the shear layer profile is ill-posed in the Sobolev settings provided that the initial velocity shear flow is non-degenerately critical at the same point.

math.AP↗

arXiv2020

Verification of Prandtl boundary layer ansatz for the steady electrically conducting fluids with a moving physical boundary

In this paper, we are concerned with the validity of Prandtl boundary layer expansion for the solutions to two dimensional (2D) steady viscous incompressible magnetohydrodynamics (MHD) equations in a domain $\{(X, Y)\in[0, L]\times\mathbb{R}_+\}$ with a moving flat boundary $\{Y=0\}$. As a direct consequence, even though there exist strong boundary layers, the inviscid type limit is still established for the solutions of 2D steady viscous incompressible MHD equations in Sobolev spaces provided that the following three assumptions hold: the hydrodynamics and magnetic Reynolds numbers take the same order in term of the reciprocal of a small parameter $ε$, the tangential component of the magnetic field does not degenerate near the boundary and the ratio of the strength of tangential component of magnetic field and tangential component of velocity is suitably small. And the error terms are estimated in $L^\infty$ sense.

math.AP↗

arXiv2019

Research on the pixel-based and object-oriented methods of urban feature extraction with GF-2 remote-sensing images

During the rapid urbanization construction of China, acquisition of urban geographic information and timely data updating are important and fundamental tasks for the refined management of cities. With the development of domestic remote sensing technology, the application of Gaofen-2 (GF-2) high-resolution remote sensing images can greatly improve the accuracy of information extraction. This paper introduces an approach using object-oriented classification methods for urban feature extraction based on GF-2 satellite data. A combination of spectral, spatial attributes and membership functions was employed for mapping the urban features of Qinhuai District, Nanjing. The data preprocessing is carried out by ENVI software, and the subsequent data is exported into the eCognition software for object-oriented classification and extraction of urban feature information. Finally, the obtained raster image classification results are vectorized using the ARCGIS software, and the vector graphics are stored in the library, which can be used for further analysis and modeling. Accuracy assessment was performed using ground truth data acquired by visual interpretation and from other reliable secondary data sources. Compared with the result of pixel-based supervised (neural net) classification, the developed object-oriented method can significantly improve extraction accuracy, and after manual interpretation, an overall accuracy of 95.44% can be achieved, with a Kappa coefficient of 0.9405, which objectively confirmed the superiority of the object-oriented method and the feasibility of the utilization of GF-2 satellite data.

cs.OH↗

arXiv2018

Novel GLHUA Electromagnetic Invisible Double Layer Cloak With Relative Parameters Not Less Than 1 and GL No Scattering Inversion,radial R in the sphere coordinate can be negative in a negative space world

Using GILD and GL no scattering modeling and inversion, we find a class of the nonzero solution of the zero scattering nonlinear inversion equation and use it to create our GLHUA cloak with relative EM parameter not less than 1 for each layer with any thickness. major new ingredients are: (1) In the outer layer, R1 < r < R2, the relative radial electric permittivity and radial magnetic permeability equal to 1; the relative angular electric permittivity and magnetic permeability are equal to 0.5((H/C)^(Alpha)+(C/H)^(Alpha)),C=R2-R1, H= r-R1; (2) In the inner layer cloak R0 < r < R1 , the relative radial electric permittivity and radial magnetic permeability equal to 1; the relative angular electric permittivity and magnetic permeability are equal to 0.5((H/C)(R0/r))^(Alpha)+((C/H)(r/R0))^(Alpha)),C=R1-R0, H= R1-r; (3) The phase velocity of wave in GLHUA cloak is less than light speed and tends to zero at r=R1;(4) We proved that GLHUA cloak is complete invisible cloak with concealment; in GLHUA cloak, the EM wave field is going to zero at r=R1. The incident wave excited in outside of GLHUA cloak can not be disturbed by cloak, and the wave can not propagation penetrate into the concealment; (5) The incident wave excited in concealment can not propagation to go to outside of the cloak, the incident wave can not be disturbed in the concealment.(6) The idea and no scattering modeling and inversion and practicable GLHUA cloak are new invisible natural science attract more researcher's effort. In my [33].we discover negative space, in which sphere radial R is negative. We found an exact full EM wave propagation solution of the EM equation in the GLHUA double layer invisible cloak, which is breakthrough progress. Analytical EM wave propagation in GLHUA cloak is published in this version. We create a novel sphere radial transformation from negative infinite to R_1 sphere radial transformation.

physics.gen-ph↗

arXiv2018

The Theoretical Proof for GLHUA EM Invisible Double Layer Cloak By Using GL No Scattering Modeling and Inversion

In this paper, we proposed the GLHUA double layer cloak; proved the properties of GLHUA double layer cloak; Using GL no scattering inversion and the pre cloak condition 6.1 to 6.4 in paper [1], we create GLHUA outer layer cloak radial relative parameter and angular relative parameter theoretically. We proved theorem 4.1 to theorem 4.4 that the phase velocity of the electromagnetic wave in GLHUA outer layer cloak is less than light speed and tends to zero at the boundary $r=R_1$, the EM wave tends to zero at $r=R1$; When source $r_s > R_2$ in the outside of the cloak and observer $r_o < R_o$ in the concealment, then $\vec E(\vec r_o)=0$ and $\vec H(\vec r_o)=0$ that is proved in theorem 5.1 and 5.3. We prove that the EM wave excited in outside of the cloak can not propagation penetrate into the concealment. In theorem 5.4 to 5.6, we rigorously proved the EM wave excited in outside of the cloak can not be disturbed by the cloak. In theorem 6.1 to theorem 6.6, we prove that the EM wave excited in the concealment can not propagate to outside of GLHUA inner cloak. we prove that the EM wave excited in the concealment can not be disturbed by the cloak. We theoretically prove that the GLHUA double cloak is invisible cloak with concealment and with relative parameter not less than 1; the GLHUA double layer cloak is practicable. In December of 2016, we submitted 3 papers to arXiv for GLHUA double layer cloak with relative parameters not less than 1. This is the third paper. The task and content of the three paper are different from each other. The second paper [1] and this Paper are theoretical base and proof of the subjects in paper arXiv:1612.02857. We find an exact analytical EM wave field of Maxwell EM equation in GLHUA cloak and mirage bridge wave. Patent of the GLHUA EM cloaks,GLHUA sphere and GL modeling and inversion methods are reserved by authors in GL Geophysical Laboratory.

physics.class-ph↗

arXiv2018

Justification of Prandtl Ansatz for MHD boundary layer

As a continuation of \cite{LXY}, the paper aims to justify the high Reynolds numbers limit for the MHD system with Prandtl boundary layer expansion when no-slip boundary condition is imposed on velocity field and perfect conducting boundary condition on magnetic field. Under the assumption that the viscosity and resistivity coefficients are of the same order and the initial tangential magnetic field on the boundary is not degenerate, we justify the validity of the Prandtl boundary layer expansion and give a $L^\infty$ estimate on the error by multi-scale analysis.

math.AP↗

arXiv2018

Global-in-time Stability of 2D MHD boundary Layer in the Prandtl-Hartmann Regime

In this paper, we prove global existence of solutions with analytic regularity to the 2D MHD boundary layer equations in the mixed Prandtl and Hartmann regime derived by formal multi-scale expansion in \cite{GP}. The analysis shows that the combined effect of the magnetic diffusivity and transveral magnetic field on the boundary leads to a linear damping on the tangential velocity field near the boundary. And this damping effect yields the global in time analytic norm estimate in the tangential space variable on the perturbation of the classical steady Hartmann profile.

math.AP↗

arXiv2018

Lifespan of Solution to MHD Boundary Layer Equations with Analytic Perturbation of General Shear Flow

In this paper, we consider the lifespan of solution to the MHD boundary layer system as an analytic perturbation of general shear flow. By using the cancellation mechanism in the system observed in \cite{LXY1}, the lifespan of solution is shown to have a lower bound in the order of $\varepsilon^{-2+}$ if the strength of the perturbation is of the order of $\varepsilon$. Since there is no restriction on the strength of the shear flow and the lifespan estimate is larger than the one obtained for the classical Prandtl system in this setting, it reveals the stabilizing effect of the magnetic field on the electrically conducting fluid near the boundary.

math.AP↗

arXiv2018

Watt-level widely tunable single-mode emission by injection-locking of a multimode Fabry-Perot quantum cascade laser

Free-running Fabry-Perot lasers normally operate in a single-mode regime until the pumping current is increased beyond the single-mode instability threshold, above which they evolve into a multimode state. As a result of this instability, the single-mode operation of these lasers is typically constrained to few percents of their output power range, this being an undesired limitation in spectroscopy applications. In order to expand the span of single-mode operation, we use an optical injection seed generated by an external-cavity single-mode laser source to force the Fabry-Perot quantum cascade laser into a single-mode state in the high current range, where it would otherwise operate in a multimode regime. Utilizing this approach we achieve single-mode emission at room temperature with a tuning range of $36 \, \mathrm{cm}^-1$ and stable continuous-wave output power exceeding 1 W. Far-field measurements show that a single transverse mode is emitted up to the highest optical power indicating that the beam properties of the seeded Fabry-Perot laser remain unchanged as compared to free-running operation.

physics.optics↗