Fractional Rényi Entropy Image Enhancement for Deep Segmentation of Kidney MRI

Recently, many rapid developments in digital medical imaging have made further contributions to health care systems. The segmentation of regions of interest in medical images plays a vital role in assisting doctors with their medical diagnoses. Many factors like image contrast and quality affect the result of image segmentation. Due to that, image contrast remains a challenging problem for image segmentation. This study presents a new image enhancement model based on fractional Rényi entropy for the segmentation of kidney MRI scans. The proposed work consists of two stages: enhancement by fractional Rényi entropy, and MRI Kidney deep segmentation. The proposed enhancement model exploits the pixel’s probability representations for image enhancement. Since fractional Rényi entropy involves fractional calculus that has the ability to model the non-linear complexity problem to preserve the spatial relationship between pixels, yielding an overall better details of the kidney MRI scans. In the second stage, the deep learning kidney segmentation model is designed to segment kidney regions in MRI scans. The experimental results showed an average of 95.60% dice similarity index coefficient, which indicates best overlap between the segmented bodies with the ground truth. It is therefore concluded that the proposed enhancement model is suitable and effective for improving the kidney segmentation performance.     

Design of sign fractional optimization paradigms for parameter estimation of nonlinear Hammerstein systems

Neural Computing and Applications - Fractional calculus plays a fundamental role in understanding the physics of nonlinear systems due to its heritage of uncertainty, nonlocality and complexity. In...     

Scattered Data Interpolation Using Cubic Trigonometric Bézier Triangular Patch

This paper discusses scattered data interpolation using cubic trigonometric Bézier triangular patches with continuity everywhere. We derive the condition on each adjacent triangle. On each triangular patch, we employ convex combination method between three local schemes. The final interpolant with the rational corrected scheme is suitable for regular and irregular scattered data sets. We tested the proposed scheme with 36,65, and 100 data points for some well-known test functions. The scheme is also applied to interpolate the data for the electric potential. We compared the performance between our proposed method and existing scattered data interpolation schemes such as Powell–Sabin (PS) and Clough–Tocher (CT) by measuring the maximum error, root mean square error (RMSE) and coefficient of determination (). From the results obtained, our proposed method is competent with cubic Bézier, cubic Ball, PS and CT triangles splitting schemes to interpolate scattered data surface. This is very significant since PS and CT requires that each triangle be splitting into several micro triangles.     

Emergent patterns in diffusive Turing-like systems with fractional-order operator

Neural Computing and Applications - Patterns obtained in abiotically homogeneous habitats are of specific interest due to the fact that they require an explanation based on the individual behavior...     

New Fuzzy Fractional Epidemic Model Involving Death Population

In this research, we propose a new change in classical epidemic models by including the change in the rate of death in the overall population. The existing models like Susceptible-Infected-Recovered (SIR) and Susceptible-Infected-Recovered-Susceptible (SIRS) include the death rate as one of the parameters to estimate the change in susceptible, infected and recovered populations. Actually, because of the deficiencies in immunity, even the ordinary flu could cause death. If people’s disease resistance is strong, then serious diseases may not result in mortalities. The classical model always assumes a closed system where there is no new birth or death, no immigration or emigration, while in reality, such assumptions are not realistic. Moreover, the classical epidemic model does not report the change in population due to death caused by a disease. With this study, we try to incorporate the rate of change in the population of death caused by a disease, where the model is framed to reduce the curve of death along with the susceptible and infected populations. Since the rate of change turned out to be very small, we have tried to estimate it fractionally. Thus, the model is defined using fuzzy logic and is solved by two different methods: a Laplace Adomian decomposition method (LADM) and a differential transform method (DTM) for an arbitrary order α. To test its accuracy, we compared the results of both DTM and LADM with the fourth-order Runge-Kutta method (RKM-4) at α = 1.     

Magnetic and dielectric properties of Co–Zn ferrite

Nanocrystalline powders of Co substituted Zn ferrite with the chemical formula Co_xZn_1−xFe₂O₄ (x = 0, 0.2, 0.4, 0.6, 0.8, 1) were synthesized by sol–gel autocombustion method using tartaric acid as fuel agent. The samples were sintered in static air atmosphere for 7 h at 773 K, 7 h at 973 K and 10 h at 1173 K. The organic phase extinction and the spinel phase formation were monitored by means of Fourier transform infrared spectroscopy. The X-ray diffraction patterns analysis confirmed the spinel single phase accomplishment. Crystallite size, average grains size, lattice parameter and cation distribution were estimated. Magnetic behavior of the as-obtained samples by means of M-H hysteresis measurements was studied at room temperature. Permeability and dielectric permittivity at room temperature versus frequency was the subject of a comparative study for the Co_xZn_1−xFe₂O₄ series. In agreement with the proposed cation distribution the sample with Co_0.8Zn_0.2Fe₂O₄ formula exhibits the optimal magnetic and dielectric properties.     

A 3D stylized half-core CANDU benchmark problem

A 3D stylized half-core Canadian deuterium uranium (CANDU) reactor benchmark problem is presented. The benchmark problem is comprised of a heterogeneous lattice of 37-element natural uranium fuel bundles, heavy water moderated, heavy water cooled, with adjuster rods included as reactivity control devices. Furthermore, a 2-group macroscopic cross section library has been developed for the problem to increase the utility of this benchmark for full-core deterministic transport methods development. Monte Carlo results are presented for the benchmark problem in cooled, checkerboard void, and full coolant void configurations.     

Analysis of a multigroup stylized CANDU half-core benchmark

An 8-group cross section library is provided to augment a previously published 2-group 3D stylized half-core Canadian deuterium uranium (CANDU) reactor benchmark problem. Reference eigenvalues and selected pin and bundle fission rates are also included. This benchmark is intended to provide computational reactor physicists and methods developers with a stylized model problem in more than two energy groups that is realistic with respect to the underlying physics. In addition to transport theory code verification, the 8-group energy structure provides reactor physicist with an ideal problem for examining cross section homogenization and collapsing effects in a full-core environment. To this end, additional 2-, 4- and 47-group full-core Monte Carlo benchmark solutions are compared to analyze homogenization-free transport approximations incurred as a result of energy group condensation.