Search

Group classification is the basic problem in the group analysis of differential equations with an arbitrary element. For the equations of ideal gas dynamics with a stationary state equation the problem is solved with a brute-force search of simplifications of the defining relations by equivalence transformations. For time-dependent relaxation state equations, the brute-force search is huge, and we have to use an optimal system of subalgebras of a subalgebra extending the core of admitted algebras. Some combination of both methods leads to a solution of the problem of group classification of gas-dynamic relaxation. [ABSTRACT FROM AUTHOR]

Background/Objectives: The auditory middle-latency responses (AMLRs) assess central sensory processing beyond the brainstem and serve as a measure of sensory gating. They have clinical relevance in the diagnosis of neurological conditions. In this study, magnitude and habituation of the AMLRs were tested for sensitivity and specificity in classifying dizzy patients with vestibular migraine (VM) and post-concussive syndrome. Methods: Twenty-three healthy individuals, 12 concussion and 26 VM patients were recruited. AMLR were recorded performing five blocks of 200 binaural click-stimulations at 60 dB sensation level with a repetition rate of 6.1 Hz. Reduction in P0, Na and Pa magnitudes between blocks was measured. Group classifications were performed through logistic and multiple regression. Results: Among healthy subjects, a consistent P0 and Na habituation can be observed. Concussed subjects show control-like Na habituation, despite a lower magnitude, while P0 habituation was negligible. VM patients showed poor habituation for all waves. Regression analyses suggest that P0 and Na better distinguish healthy subjects from neurological patients, whereas Pa best distinguishes concussion from VM patients. Conclusions: The results support that AMLR habituation can contribute to unraveling different mechanisms of dizziness due to concussion compared to VM, providing insights that can complement routine diagnostic assessments. [ABSTRACT FROM AUTHOR]

In this paper, we present Lie symmetry analysis for time fractional Black-Scholes equation with time-dependent coefficients. The group classification is carried out by investigating the time-dependent coefficients σ (t) , r(t) and s(t). Then the obtained group generators are used to reduce the equation under study, some of the reduced equations are fractional ordinary equations with Erdélyi-Kober fractional derivative, and some exact solutions including power series solutions are constructed. [ABSTRACT FROM AUTHOR]

Unsteady equations of flat and axisymmetric boundary layers are considered. For the unsteady axisymmetric boundary layer equation, the problem of group classification is solved. It is shown that the kernel of symmetry operators can be extended by no more than four-dimensional Lie algebra. The kernel of symmetry operators of the unsteady flat boundary layer equation is found and it is shown that it can be extended by no more than a five-dimensional Lie algebra. The non-existence of the unsteady analogue of the Stepanov–Mangler transformation is proved. [ABSTRACT FROM AUTHOR]

The group analysis of differential equations of ideal gas dynamics is most developed. Earlier, the state equations for thermodynamic parameters were assumed to be time-independent. The time dependence may take place for relaxing media, for example, as a result of rheology or due to the energy averaging of processes in a multiphase medium. The problem of group analysis of relaxing media is posed. First, equivalence transformations are calculated that change only the state equations. Next, the problem of group classification is solved: it is required to find, up to equivalence transformations, classes of state equations for which the admitted group is extended. This problem is partially solved in the present paper. [ABSTRACT FROM AUTHOR]

In this paper, we investigate the time-fractional generalized thin film equation (TFGTFE) with two arbitrary functions and perform the group classification with respect to these arbitrary functions. Specifically, all vector fields admitted by the considered equations are provided utilizing Lie symmetry analysis. Then the corresponding symmetry reductions are carried out and exact solutions to some special equations are obtained. In particular, we construct the power series solutions to one type of TFGTFE by means of the combination of the Erd e ´ lyi-Kober (EK) operator with the analytic power series method and verify the convergence of the power series solutions using implicit function theorem. In addition, taking advantage of Matlab software, the three-dimensional diagrams and two-dimensional graphs of some obtained solutions are demonstrated for the purpose of visualization. [ABSTRACT FROM AUTHOR]

In this paper, we perform a detailed group classification for a generalized convection-reaction-diffusion equation with three unknown functions. Specifically, we determine all the functional forms for the unknown functions where the given equation admits nontrivial Lie point symmetries. The classification problem provides us with eight families of equations summarized in four categories. The admitted Lie symmetries form the four Lie algebras 2 A 1 , A 4 , 4 , A 2 , 1 ⊕ A 1 and A 2 , 1 ⊕ A 2 , 1 . For the four families of the classification problem we calculate the one-dimensional optimal system and we derive all the similarity transformations which reduce the partial differential equation into an ordinary differential equation. Applications of the similarity transformations are presented while exact solutions are derived. [ABSTRACT FROM AUTHOR]

We solve the main problems of group analysis for differential equations of ideal gas dynamics on assuming that the state equation for thermodynamic parameters is independent of time. Using group analysis, we study a relaxing medium that changes with time, for example, as a result of rheology or due to energy averaging of processes in a multiphase medium. Equivalence transformations change the state equation rather than the motion equation. The computations of equivalence transformations define some preliminary group classification, i.e., the classes of state equations such that the group of equivalence transformations changes. We show that projective transformations apply to more general than stationary state equations. Also, we propose some new constructive method for group classification by calculating equivalence transformations with additional conditions on the state equations that appear while analyzing the invariance conditions. [ABSTRACT FROM AUTHOR]

We consider a general class of systems of three partial differential equations and we provide restrictions on the form of Lie symmetry operators admitted by such systems. When these restrictions are known in advance, the symmetry analysis becomes simpler. Special cases of this class are two generalizations of Hirota–Satsuma systems with variable coefficients. We derive the equivalence groups for these two systems. With the aid of the equivalence groups and the restricted form of the Lie symmetry operators, we present an enhanced Lie group classification for the two Hirota–Satsuma systems. For each class a specific system is single out which has the property to be mapped into a constant coefficient system.

A fractional order system of evolution partial differential equations with a constant delay is considered. By exploiting the Lie symmetry method, we give a complete group classification of the system. Furthermore, we establish the corresponding symmetry reductions and construct some analytical solutions to the system. [ABSTRACT FROM AUTHOR]

The two-dimensional Green–Naghdi equations are studied for the case of uneven bottom topography. The bottom topography function can depend on time. Group classification of these equations with respect to the function describing the bottom topography is performed using an algebraic approach. [ABSTRACT FROM AUTHOR]

The attributes of the knowledge nodes in the interactive educational knowledge graph need to cater to students' online learning preferences, so understanding the composition and learning preferences of students in the online learning process is helpful to the development of more targeted learning paths. Currently, there are few existing research results on knowledge graph embedding methods based on students' interaction with respect to knowledge points, student group composition and their learning preferences. To this end, this paper studies the construction of an interactive educational knowledge graph and the classification of student groups. First, a knowledge recommendation idea was proposed based on the classification of student groups. Through the three types of interaction behaviors -- human-computer, teacher-student, and student-student interactions that occur on the online learning platform, the depth of students' interactions with respect to the knowledge points in the interactive educational knowledge graph was characterized. The online learning effects of students were quantified by the interactive achievement of knowledge points mastered by students and the weights of knowledge points which represent their importance. Then, the effects of the differences in the interactions of students with respect to different knowledge points on the stability of the similarity prediction of students' learning preferences were explored, and based on the analysis results, students were classified into groups. The experimental results verified the effectiveness of the proposed model. [ABSTRACT FROM AUTHOR]

As one of the pillar industries of China's national economy the clothing and textile industry creates huge foreign exchange but also causes serious environmental pollution so it should bear the responsibility of reducing carbon dioxide emissions. On October 28 2021 National Development and Reform Commission made important arrangements for the goal of achieving peak carbon emissions by 2030 which provided a clear guiding plan for the reform of the economic and social development in China and the green transformation of the clothing and textile industry is imminent. The achievement of the goal of achieving peak carbon emissions of the clothing and textile industry requires the green development of the entire supply chain and the formation of a low-carbon supply chain system. However although the clothing and textile industry in China has mastered certain sustainable development competitiveness in the upstream and midstream of the supply chain there are still many unsustainable phenomena on the consumer side due to the lack of targeted sustainable concept guidance. In order to guide consumers to develop the concept of sustainable clothing consumption the study builds portraits of Chinese clothing consumers based on the CART algorithm to understand the basic characteristics of sustainable clothing consumers. The study designs indicators for profile construction from demographic variables and sustainable living habits. In order to get the category of sustainable clothing consumers first of all the sustainable clothing consumption behaviors are divided into purchase use and disposal stages and 1 835 consumers are measured in the three stages of sustainable clothing consumption behavior through a 5-point digital scale. Then based on the quantitative score of sustainable behavior K-means algorithm is adopted to divide sustainable clothing consumers into " low-active type" " mid-active type" and "active type" . In order to build "mid-active type" and "active type" consumers, portraits " low-active type" consumers are taken as a control group to filter and combine portrait indicators. First of all the chi-square test and Logistic regression test are used to screen the portrait indicators which indicates that urban-rural differences and marital status are not included in the regression model of consumer attributes. Then the CART algorithm is used to construct "mid-active type" and "active type" sustainable clothing consumer portraits which not only extracts the main factors that affect the construction of consumer portraits but also ranks the importance of the main factors. It is found that living habits intergeneration and gender are important indicators for building " mid-active type" sustainable clothing consumer portraits. Consumers who actively classify garbage and purchase green food are likely to become "mid-active type" sustainable clothing consumers and the "post-95s generation" and female are also likely to become "mid-active type" sustainable clothing consumers. It is also found that living habits the status and number of children and educational background are important indicators for building "active type" sustainable clothing consumer portraits. Consumers who have purchased new energy vehicles/ low-carbon homes are likely to become " active type" sustainable clothing consumers and consumers with many children and consumers with low education are also likely to become "active type" sustainable clothing consumers. By building the portrait of sustainable clothing consumers in China it is conducive to in-depth understanding of the internal mechanism of sustainable clothing consumption behavior. In order to achieve sustainable development in the field of clothing consumption under the policy of achieving peak carbon emissions it is essential to pay attention to many aspects such as the publicity of overall environmental behavior the promotion of family environmental awareness the education of sustainable clothing behavior the implementation of economic subsidies for green clothing and so on. The research results can provide theoretical support for the consumer segmentation of sustainable clothing consumption and the practice of sustainable consumption behavior under the policy of achieving peak carbon emissions and carbon neutrality. [ABSTRACT FROM AUTHOR]

The Guéant and Pu model of option pricing and hedging, which takes into account transaction costs, and the impact of operations on the market is studied by group analysis methods. The infinite-dimensional continuous group of equivalence transforms of the model is found. It is applied to get the group classification of the model under consideration. In addition to the general case, the classification contains three specifications of a free element in the equation, which correspond to models with groups of symmetries of a special kind. Optimal systems of subalgebras for some concrete models from the obtained classification are derived and used for the calculation of according invariant submodels. [ABSTRACT FROM AUTHOR]

Lie symmetries of Schrödingerâ€"Pauli equations for charged particles and quasirelativistic Schrödinger equations are classified. In particular a new superintegrable system with spinâ€"orbit coupling is discovered. [ABSTRACT FROM AUTHOR]

A nonlinear two-dimensional orthotropic filtration equation with the Riemann–Liouville time-fractional derivative is considered. It is proved that this equation can admits only linear autonomous groups of point transformations. The Lie point symmetry group classification problem for the equation in question is solved with respect to coefficients of piezoconductivity. These coefficients are assumed to be functions of the square of the pressure gradient absolute value. It is proved that if the order of fractional differentiation is less than one then the considered equation with arbitrary coefficients admits a four-parameter group of point transformations in orthotropic case, and a five-parameter group in isotropic case. For the power-law piezoconductivity, the group admitted by the equation is five-parametric in orthotropic case, and six-parametric in isotropic case. Also, a special case of power function of piezoconductivity is determined for which there is an additional extension of admitted groups by the projective transformation. There is no an analogue of this case for the integer-order filtration equation. It is also shown that if the order of fractional differentiation $\alpha \in (1,2)$ then dimensions of admitted groups are incremented by one for all cases since an additional translation symmetry exists. This symmetry is corresponded to an additional particular solution of the fractional filtration equation under consideration. Using the group classification results for orthotropic case, the representations of group-invariant solutions are obtained for two-dimensional subalgebras from optimal systems of symmetry subalgebras. Examples of reduced equations obtained by the symmetry reduction technique are given, and some exact solutions of these equations are presented. It is proved that the considered time-fractional filtration equation is nonlinearly self-adjoint and therefore the corresponding conservation laws can be constructed. The components of obtained conserved vectors are given in an explicit form.

The system of equations describing the dynamics of non-isothermal mixture of two ideal gases is considered. It contains a free element, which depends on internal energies of phases. For this system the Lie algebra of the equivalence transformations group is found, the principal Lie algebra of the system is obtained. A group classification problem with respect the free element is solved. The obtained symmetries are used for the search of conservation laws of the equations by the method of nonlinear self-adjointness. The optimal system of subalgebras of the symmetry algebra is constructed. The search for invariant and partially invariant solutions and submodels has been carried out for all subalgebras from the optimal system. [ABSTRACT FROM AUTHOR]

Copyright of Xinjiang Agricultural Sciences is the property of Xinjiang Agricultural Sciences Editorial Department and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

We consider a system of partial differential equations modeling tumors. The system under consideration describes the spatial dynamics of the tumor cells, extracellular matrix, and matrix degrading enzymes. We first carry out a complete group classification of the Lie point symmetries of this model. Next, we use symmetry techniques to construct invariant solutions for it. In addition, we consider a second system of partial differential equations, coupling to the original one the concentration of oxygen, and we find several analytical solutions to this system. Most of the solutions are biologically relevant and consistent with the evolution of such tumors. [ABSTRACT FROM AUTHOR]

Drivers’ route switching behavior shows obvious difference when they face various traffic conditions. The article studies the drivers’ route switching behavior based on group classification. Questionnaire combining SP survey and RP survey is carried out to collect the drivers’ route choice behavior under the influence of individual attributes, daily travel characteristic and traffic conditions. Latent Class Model (LCM) is used to analyze the behavior characteristic. According to the goodness of models, drivers are divided into three categories. Drivers of sensitive pattern will switch route easily which is represented by young people with shorter driver-age. In contrast, drivers of unresponsive pattern will not switch routes easily and the pattern is represented by elder people with longer driver-age. Based on the drivers’ classification results, ordinal logistic model is established. According to the odds ratio of each variable, we find that, age, driver-age, nature of drivers, and travel frequency all affect drivers’ route switching behavior.

A group analysis of the second-order equation including the one-dimensional gas dynamics equations in Lagrangian coordinates as a particular case is performed. The use of Lagrangian coordinates makes it possible to consider the one-dimensional gas dynamics equations as a variational Euler-Lagrange equation with an appropriate Lagrangian. Conservation laws are derived with the use of the variational presentation and Noether's theorem. A complete group classification of the Euler-Lagrange equation is obtained; as a result, 18 different classes can be identified. [ABSTRACT FROM AUTHOR]

In the application of statistical methods and models, as well as other data analysis methods, we encounter a number of limitations that make it necessary to find new approaches. The purpose of this article is to highlight the rationale for seeking new approaches to the analysis of mortality data in European countries observed by sex, age, and cause of death, to briefly present some results of recent work, and to point to research being conducted in collaboration with other researchers on this information.

• Flows of one-dimensional continuum in Lagrangian coordinates are studied. • Lagrangian coordinates used to find new conservation laws. • Using Noether's theorem, conservation laws are found. The equations describing the flow of a one-dimensional continuum in Lagrangian coordinates are studied in this paper by the group analysis method. They are reduced to a single Euler–Lagrange equation which contains two undetermined functions (arbitrary elements). Particular choices of these arbitrary elements correspond to different forms of the shallow water equations, including those with both, a varying bottom and advective impulse transfer effect, and also some other motions of a continuum. A complete group classification of the equations with respect to the arbitrary elements is performed. One advantage of the Lagrangian coordinates consists of the presence of a Lagrangian, so that the equations studied become Euler–Lagrange equations. This allows us to apply Noether's theorem for constructing conservation laws in Lagrangian coordinates. Not every conservation law in Lagrangian coordinates has a counterpart in Eulerian coordinates, whereas the converse is true. Using Noether's theorem, conservation laws which can be obtained by the point symmetries are presented, and their analogs in Eulerian coordinates are given, where they exist. [ABSTRACT FROM AUTHOR]

A problem of the Lie point approximate symmetry group classification of a perturbed subdiffusion equation with a small parameter is solved. The classification is performed with respect to anomalous diffusion coefficient which is considered as a function of an independent variable. The perturbed subdiffusion equation is derived from a fractional subdiffusion equation with the Riemann-Liouville time-fractional derivative under an assumption that the order of fractional differentiation is close to unity. As it is follow from the classification results, the perturbed subdiffusion equation admits a more general Lie point symmetry group than the initial fractional subdiffusion equation. The obtained results permit to construct approximate invariant solutions for the perturbed subdiffusion equation corresponding to different functions of the anomalous diffusion coefficient. These solutions will also be the approximate solutions of the initial fractional subdiffusion equation.

We study generalized Kompaneets equations (GKEs) with one functional parameter, and using the Lie-Ovsiannikov algorithm, we carried out the group classification. It is shown that the kernel algebra of the full groups of the GKEs is the one-dimensional Lie algebra. Using the direct method, we find the equivalence group. We obtain six non-equivalent (up to transformations from the equivalence group) GKEs that allow wider invariance algebras than the kernel one. We find a number of exact solutions of the non-linear GKE which has the maximal symmetry properties.

A general nonlinear model of illiquid markets with feedback effects is considered. This equation with 2 free functional parameters contains as partial cases the classical Black‐Scholes equation, Schönbucher‐Wilmott equation, and Sircar‐Papanicolaou equation of option pricing. We obtain here the complete group classification of the equation, and for every parameters specification we obtain the principal Lie algebra and its optimal system of 1‐dimensional subalgerbras. For every such subalgebra we calculate the invariant submodel and invariant solution, when it is possible. Thus, the series of invariant submodels and invariant solutions are derived for the considered nonlinear model. [ABSTRACT FROM AUTHOR]

In this paper, the group classification is presented for Fisher equation with time-dependent coefficients. The analysis provided many new solutions that take the form of doubly periodic-like solutions and soliton-like solutions. [ABSTRACT FROM AUTHOR]

A group classification is preformed on a sixth-order thin film type equation under the group of continuous equivalence transformation. A list of similar reductions is presented based on the optimal systems of one-dimensional subalgebras of the corresponding symmetry algebras. Some invariant solutions with physical interest are constructed. The nonlinear self-adjointness of the equation is showed and conservation laws are obtained by using the new conservation theorem proposed by Ibragimov. [ABSTRACT FROM AUTHOR]

Abstract: Faba bean (Vicia faba L.) is an important source of protein, but breeding for increased yield stability and stress tolerance is hampered by the scarcity of phenotyping information. Because comparisons of cultivars adapted to different agroclimatic zones improve our understanding of stress tolerance mechanisms, the root architecture and morphology of 16 European faba bean cultivars were studied at maturity. Different machine learning (ML) approaches were tested in their usefulness to analyse trait variations between cultivars. A supervised, that is, hypothesis‐driven, ML approach revealed that cultivars from Portugal feature greater and coarser but less frequent lateral roots at the top of the taproot, potentially enhancing water uptake from deeper soil horizons. Unsupervised clustering revealed that trait differences between northern and southern cultivars are not predominant but that two cultivar groups, independently from major and minor types, differ largely in overall root system size. Methodological guidelines on how to use powerful ML methods such as random forest models for enhancing the phenotypical exploration of plants are given. [ABSTRACT FROM AUTHOR]

A one-dimensional nonlinear fractional filtration equation with the Riemann-Liouville time-fractional derivative is proposed for modeling fluid flow through a porous medium. This equation is derived under an assumption that the fluid has a fractional equation of state in which the fluid density depends on the time-fractional derivative of pressure. The obtained equation belongs to the diffusion-wave type of equations. A case when the order of fractional differentiation is close to an integer number is considered, and a small parameter is introduced into the fractional filtration equation under consideration. An expansion of the Riemann-Liouville time-fractional derivative into the series with respect to this small parameter is obtained. Using this expansion, a first-order approximation of the derived fractional filtration equation is performed. Next, the problem of approximate Lie point symmetry group classification for this approximate nonlinear filtration equation with a small parameter is studied. It is shown that approximate symmetry groups admitted by different realizations of the approximate filtration equation have much more dimensions than the corresponding exact Lie point symmetry groups admitted by unperturbed fractional diffusion-wave equations. Obtained classification results permit to construct approximate invariant solutions for the considered nonlinear time-fractional filtration equations. [ABSTRACT FROM AUTHOR]

This paper describes the solution of group classification problem for heat and mass transfer equations with respect to 3 transport coefficients. Two coefficients depend on temperature and concentration, and the thermal diffusivity coefficient is the function of only one of these state parameters. The forms of the arbitrary elements providing the additional transformations are found. Examples of exact solutions of the governing equations are constructed. [ABSTRACT FROM AUTHOR]

Lie group classification for a diffusion‐type system that has applications in plasma physics is derived. The classification depends on the values of 5 parameters that appear in the system. Similarity reductions are presented. Certain initial value problems are reduced to problems with the governing equations being ordinary differential equations. Examples of potential symmetries are also presented. [ABSTRACT FROM AUTHOR]

Lie group classification for 2 Burgers-type systems is obtained. Systems contain 2 arbitrary elements that depend on the 2 dependent variables. Equivalence transformations for the systems are derived. Examples of nonclassical reductions are given. A Hopf-Cole--type mapping that linearizes a nonlinear systemis presented. [ABSTRACT FROM AUTHOR]

The study of cognitive processes performed by the human brain has greatly benefited from new technologies able to infer neuronal activity by means of noninvasive methods. This is the case of functional magnetic resonance imaging. Digital image analysis and interpretation techniques have also contributed greatly to the exploration and understanding of these brain functions. Among these techniques, the use of machine learning algorithms with the ability to automatically classify cognitive states has been particularly fruitful. In general terms, these techniques identify brain regions involved in specific cognitive processes by correlating experimental stimulation patterns with the magnitude of observed neural activity. An issue with this kind of analyses arises when comparing activation results from different subjects. This occurs due to functional and anatomical variability among individuals, even after this variability is reduced during the registration process performed on the images as part of the preprocessing. In this paper we propose a feature selection method to contend with this variability. The basic idea consists in defining the activity of a voxel (feature) as a weighted vote of the observed activity of its neighbors located at a periphery defined by an isotropic three-dimensional space. Such influence is determined by a Gaussian radial function. This approach allows comparing results among different individuals, assuming that functionally equivalent activities are not necessarily presented in the same spatial position. The results show that this spatial tolerance allows a classification accuracy of 96% (considering a threshold of ± 2 voxels, equivalent to ± 8 mm.) against the 84% obtained by a traditional feature selection method. [ABSTRACT FROM AUTHOR]

Differential-difference equations of the form u⃛ = F ( t, u , u , u , u̇ , u̇ , u̇ ) are classified according to their intrinsic Lie point symmetries, equivalence group and some low-dimensional Lie algebras including the Abelian symmetry algebras, nilpotent nonAbelian symmetry algebras, solvable symmetry algebras with nonAbelian nilradicals, solvable symmetry algebras with Abelian nilradicals and nonsolvable symmetry algebras. Here F is a nonlinear function of its arguments and the dot over u denotes differentiation with respect to t. [ABSTRACT FROM AUTHOR]

Group classification of a class of systems of diffusion equations is carried out. Arbitrary elements that appear in the system depend on two variables. All forms of the arbitrary elements that provide additional Lie symmetries are determined. Equivalence transformations are used to simplify the analysis. Examples of similarity reductions are presented. Copyright © 2016 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]

• Group classification performed with the new method of indeterminates. • Soliton solutions and other solutions found by four different methods. • General results derived on properties of four order Lagrange equations. We apply an extension of a new method of group classification to a family of nonlinear wave equations labelled by two arbitrary functions, each depending on its own argument. The results obtained confirm the efficiency of the proposed method for group classification, termed the method of indeterminates. A model equation from the classified family of fourth order Lagrange equations is singled out. Travelling wave solutions of the latter are found through a similarity reduction by variational symmetry operators, followed by a double order reduction into a second order ordinary differential equation. Multi-soliton solutions and other exact solutions are also found by various methods including Lie group and Hirota methods. The most general action of the full symmetry group on any given solution is provided. Some remarkable facts on Lagrange equations emerging from the whole study are outlined. [ABSTRACT FROM AUTHOR]

The present study explores if EEG spectral parameters can discriminate between healthy elderly controls (HC), Alzheimer's disease (AD) and vascular dementia (VaD) using. We considered EEG data recorded during normal clinical routine with 114 healthy controls (HC), 114 AD, and 114 VaD patients. The spectral features extracted from the EEG were the absolute delta power, decay from lower to higher frequencies, amplitude, center and dispersion of the alpha power and baseline power of the entire frequency spectrum. For discrimination, we submitted these EEG features to regularized linear discriminant analysis algorithm with a 10-fold cross-validation. To check the consistency of the results obtained by our classifiers, we applied bootstrap statistics. Four binary classifiers were used to discriminate HC from AD, HC from VaD, AD from VaD, and HC from dementia patients (AD or VaD). For each model, we measured the discrimination performance using the area under curve (AUC) and the accuracy of the cross-validation (cv-ACC). We applied this procedure using two different sets of predictors. The first set considered all the features extracted from the 22 channels. For the second set of features, we automatically rejected features poorly correlated with their labels. Fairly good results were obtained when discriminating HC from dementia patients with AD or VaD (AUC = 0.84). We also obtained AUC = 0.74 for discrimination of AD from HC, AUC = 0.77 for discrimination of VaD from HC, and finally AUC = 0.61 for discrimination of AD from VaD. Our models were able to separate HC from dementia patients, and also and to discriminate AD from VaD above chance. Our results suggest that these features may be relevant for the clinical assessment of patients with dementia. [ABSTRACT FROM AUTHOR]