Consensus of Feature Selection Methods and Reduced Generalization Gap Model to Improve Diagnosis of Heart Disease

Enhancing the diagnostic ability of Machine Learning models for acceptable prediction in the healthcare community is still a concern. There are critical care disease datasets available online on which researchers have experimented with a different number of instances and features for similar disease prediction. Further, different Machine Learning (ML) models have different preprocessing requirements. Framingham heart disease data is multicollinear and has missing values. Thus, the proposed model aims to explore the differential preprocessing needs of ML models followed by feature selection in consensus with domain experts and feature extraction to resolve multicollinearity issues. Missing values have been imputed differently for each feature. The work also identifies optimal train set size by plotting a learning curve that provides a minimum generalization gap. When testing is done on this hyperparameter tuned model, performance is enhanced with respect to the F score weighted by support and stratification since the data is imbalanced. Experimental results demonstrate improvement in performance metrics, i.e., weighted F score, precision, recall, accuracy up to 3 %, and F1 score by 8 % for Logistic Regression Classifier with the proposed model. Further, the time required for hyperparameter tuning is reduced by 50% for tree-based models, particularly Classification and Regression Tree (CART).

MHD Boundary Layer Flow of Casson Fluid Past an Inclined Plate in the Presence of Soret/Dufour Effects, Heat Source and First-order Chemical Reaction

The main objective of this study is to investigate the effects of the Casson fluid parameter on an incompressible, magnetohydrodynamic boundary layer flow of a Casson fluid past a moving porous inclined plate in the presence of heat source and first-order chemical reaction. The governing partial differential equations are converted into ordinary differential equations using similarity transformation and then are solved numerically, adopting bv4pc method. The effects of relevant parameters on the velocity, temperature and concentration profiles are analyzed graphically. Also, tabular form is used to present skin friction, heat transfer and mass transfer. This investigation reveals that the Casson fluid parameter enhances the fluid velocity, skin friction and Sherwood number, while the Nusselt number decreases.

Study of Zero Temperature Ground State Properties of the Repulsive Bose-Einstein Condensate in an Anharmonic Trap

The zero-temperature ground state properties of experimental 87Rb condensate are studied in a harmonic plus quartic trap [ V(r) = ½mω2r2 + λr4 ]. The anharmonic parameter (λ) is slowly tuned from harmonic to anharmonic. For each choice of λ, the many-particle Schrödinger equation is solved using the potential harmonic expansion method and determines the lowest effective many-body potential. We utilize the correlated two-body basis function, which keeps all possible two-body correlations. The use of van der Waals interaction gives realistic pictures. We calculate kinetic energy, trapping potential energy, interaction energy, and total ground state energy of the condensate in this confining potential, modelled experimentally. The motivation of the present study is to investigate the crucial dependency of the properties of an interacting quantum many-body system on λ. The average size of the condensate has also been calculated to observe how the stability of repulsive condensate depends on anharmonicity. In particular, our calculation presents a clear physical picture of the repulsive condensate in an anharmonic trap.

Comparative Analysis on Deep Learning Approaches for Heavy-Vehicle Detection based on Data Augmentation and Transfer-Learning techniques

Vehicle detection and recognition require demanding advanced computational intelligence and resources in a real-time traffic surveillance system for effective traffic management of all possible contingencies. One of the focus areas of deep intelligent systems is to facilitate vehicle detection and recognition techniques for robust traffic management of heavy vehicles. The following are such sophisticated mechanisms: Support Vector Machine (SVM), Convolutional Neural Networks (CNN), Regional Convolutional Neural Networks (R-CNN), You Only Look Once (YOLO) model, etcetera. Accordingly, it is pivotal to choose the precise algorithm for vehicle detection and recognition, which also addresses the real-time environment. In this study, a comparison of deep learning algorithms, such as the Faster R-CNN, YOLOv2, YOLOv3, and YOLOv4, are focused on diverse aspects of the features. Two entities for transport heavy vehicles, the buses and trucks, constitute detection and recognition elements in this proposed work. The mechanics of data augmentation and transfer-learning is implemented in the model; to build, execute, train, and test for detection and recognition to avoid over-fitting and improve speed and accuracy. Extensive empirical evaluation is conducted on two standard datasets such as COCO and PASCAL VOC 2007. Finally, comparative results and analyses are presented based on real-time.

A Short Review on Heterocyclic Compounds Showing Anti-Breast Cancer Activity

Breast cancer (BC) is the most common cancer for females, and its incidence tends to increase year by year. Currently, the backbone of therapy for BC is mainly chemotherapy; however, its toxicity in normal cells and acquired tumor resistance to the drug users are considered the main barriers. Therefore, there is still an urgent need for the development of more effective and safer anti-BC agents. Based on previous reference documents in recent years, this review covers the work reported on the anti-BC compounds classified according to the structures. This review summarized significant anti-BC compounds organized by functional groups according to the animal model data, although there would be some limitations. This review highlights the properties of new compounds having promising anti-BC properties, which may be proven to be more effective and selective, and possibly free of unwanted side effects. The reviewed compounds represent an interesting possibility of overcoming BC and reducing the percentage of patients with an inadequate response to drug therapy.

The Optical Properties of Dy3+ Doped LaPO4 Phosphors Prepared by Solid State Reaction Method

Luminescent nanomaterials are used in everyday life due to their employment in distinct fields of science and technology, like cathode ray tubes (CRTs), flat panel display devices, temperature sensors, lasers, solar-cells, biological imaging, and solid-state lighting but also as carriers for miscellaneous therapeutic drugs. We have prepared dysprosium Dy3+ (0.5 mol %) doped lanthanum phosphate (LaPO4) phosphor by solid state reaction method. The excitation spectra of synthesized phosphor at 595 nm monitoring were composed of broadband and a series of sharp peaks, the strongest excitation peak at 254, 271, and 350 nm. The main emission spectra of samples under 254, 271, and 350 nm excitation are Dy3+ (0.5 mol %) doped LaPO4 phosphor observed at 477 and 573 nm corresponding to blue and yellow color. The broadband emission is the characteristic of the allowed f-h transition of Dy3+ ions. The corresponding emission band is observed due to the (blue emission) 4f9/2→6h15/2, (yellow emission) 4f9/2→6h13/2 transition of Dy3+ ions. All the samples have been characterized by X-ray powder diffraction (XRD), scanning electron microscopy (SEM), Fourier transform infrared spectroscopy (FTIR), and photoluminescence (PL) techniques.

Productivity and Profitability Estimates for Sonali Chicken Farms in Rajshahi, Bangladesh

Season- and farm-wise productivity and mortality of Sonali chickens in 53 selected poultry farms from nine Upozillas of Rajshahi District during December 2018 and November 2019 were assessed. Results revealed that, on average, small farms produced 775, medium farms 1828 and large farms 3442 marketable live birds. Mortality was recorded in the following order: small farms > medium farms > large farms. Highest number of birds was produced in spring followed by winter, summer and rainy season, whereas the highest mortality was recorded in winter followed by rainy, summer and spring. Birds reared in smaller farms consumed greater amount of feed compared to those reared in larger farms. Consequently, the live weights, edible weights and edible ratios of the chickens differed significantly due to the farm size. Season-wise variations were significant for day-old chick price, gross return per bird and benefit-cost ratio. Farm-wise variations in the profitability components demonstrated that all the components of the large farms were significantly higher than those of the medium and small farms. With regard to the commercial poultry enterprise of the country, therefore, the present findings on Sonali chicken farming in Rajshahi District are quite encouraging from productivity, profitability and sustainability points of views.

Modeling of Nonlinear Variable Viscosity on Peristaltic Transport of Fluid with Slip Boundary Conditions: Application to Bile Flow in Duct

The present article deals with variable viscosity on the peristaltic transport of bile in an inclined duct under the action of slip boundary conditions. The wall geometry is described by the sinusoidal wave propagating in the axial direction with different amplitude and with constant speed. The flow of fluid is examined in a wave frame of reference, moving with the velocity of the wave. Mathematical modeling of the problem includes equations of motion and continuity. The fluid flow is investigated by converting the equations into a non-dimensionalized form simplified considering long wavelength and low Reynolds number approximation. The analytic expressions for axial velocity, pressure gradient, and pressure rise over a single wavelength cycle are obtained. The impact of various parameters such as slip parameter, viscosity parameter, angle of inclination, gravity parameter and amplitude ratio on axial velocity, pressure gradient and pressure rise are discussed in detail by plotting graphs in MATLAB R2018b software. In this article, a comparison of linear and nonlinear variation of viscosity of bile has been made. It is concluded that velocity and pressure rise is more in case linear variation of viscosity, whereas more pressure gradient is required in case of nonlinear variation of viscosity.

A New Computational Approach for Solving Fractional Order Telegraph Equations

In this work, a modified decomposition method namely Sumudu-Adomian Decomposition Method (SADM) is implemented to find the exact and approximate solutions of fractional order telegraph equations. The derivatives of fractional-order are expressed in terms of caputo operator. Some numerical examples are illustrated to examine the efficiency of the proposed technique. Solutions of fractional order telegraph equations are obtained in the form of a series solution. It is observed that the solutions of fractional order telegraph equations converge towards the solution of an integer-order problem, which confirmed the reliability of the suggested method.

Reconstruction of the New Agegraphic Polytropic Gas Dark Energy Model in the Flat Friedmann Robertson Walker (FRW) Universe

In this paper, we study the Polytropic Gas Dark Energy model and New Agegraphic Dark Energy model in the flat Friedmann Robertson Walker (FRW) Universe and establish a correspondence between them for the scalar fields. This correspondence allows reconstructing the potential of the Polytropic Gas scalar fields and dynamics of the scalar fields according to the evolutions of the New Agegraphic Dark Energy, which describes the accelerated expansion of the Universe.