Computation

15 pages, 3344 KiB

Open AccessArticle

Waiting Time Control Chart for M/G/1 Retrial Queue

by Yih-Bey Lin, Tzu-Hsin Liu, Yu-Cheng Tsai and Fu-Min Chang

Computation 2024, 12(9), 191; https://doi.org/10.3390/computation12090191 - 19 Sep 2024

Retrial queues are used extensively to model many practical problems in service systems, call centers, data centers, and computer network systems. The average waiting time is the main observable characteristic of the retrial queues. Long queues may cause negative impacts such as waste [...] Read more.

Retrial queues are used extensively to model many practical problems in service systems, call centers, data centers, and computer network systems. The average waiting time is the main observable characteristic of the retrial queues. Long queues may cause negative impacts such as waste of manpower and unnecessary crowding leading to suffocation, and can even cause trouble for customers and institutions. Applying control chart technology can help managers analyze customers’ waiting times to improve the effective performance of service and attention. This paper pioneers the developing and detailed study of a waiting time control chart for a retrial queue with general service times. Two waiting time control charts, the Shewhart control chart, and a control chart using the weighted variance method are constructed in this paper. We present three cases for the Shewhart control chart in which the service time obeys special distributions, such as exponential, Erlang, and hyper-exponential distributions. The case of an exponentially distributed service time is also presented for the control chart using the weighted variance method. Based on the numerical simulations conducted herein, managers can better monitor and analyze the customers’ waiting times for their service systems and take preventive measures. Full article

► Show Figures

Figure 1

18 pages, 1556 KiB

Open AccessArticle

Bayesian Optimized Machine Learning Model for Automated Eye Disease Classification from Fundus Images

by Tasnim Bill Zannah, Md. Abdulla-Hil-Kafi, Md. Alif Sheakh, Md. Zahid Hasan, Taslima Ferdaus Shuva, Touhid Bhuiyan, Md. Tanvir Rahman, Risala Tasin Khan, M. Shamim Kaiser and Md Whaiduzzaman

Computation 2024, 12(9), 190; https://doi.org/10.3390/computation12090190 - 16 Sep 2024

Abstract

Eye diseases are defined as disorders or diseases that damage the tissue and related parts of the eyes. They appear in various types and can be either minor, meaning that they do not last long, or permanent blindness. Cataracts, glaucoma, and diabetic retinopathy [...] Read more.

Eye diseases are defined as disorders or diseases that damage the tissue and related parts of the eyes. They appear in various types and can be either minor, meaning that they do not last long, or permanent blindness. Cataracts, glaucoma, and diabetic retinopathy are all eye illnesses that can cause vision loss if not discovered and treated early on. Automated classification of these diseases from fundus images can empower quicker diagnoses and interventions. Our research aims to create a robust model, BayeSVM500, for eye disease classification to enhance medical technology and improve patient outcomes. In this study, we develop models to classify images accurately. We start by preprocessing fundus images using contrast enhancement, normalization, and resizing. We then leverage several state-of-the-art deep convolutional neural network pre-trained models, including VGG16, VGG19, ResNet50, EfficientNet, and DenseNet, to extract deep features. To reduce feature dimensionality, we employ techniques such as principal component analysis, feature agglomeration, correlation analysis, variance thresholding, and feature importance rankings. Using these refined features, we train various traditional machine learning models as well as ensemble methods. Our best model, named BayeSVM500, is a Support Vector Machine classifier trained on EfficientNet features reduced to 500 dimensions via PCA, achieving 93.65 ± 1.05% accuracy. Bayesian hyperparameter optimization further improved performance to 95.33 ± 0.60%. Through comprehensive feature engineering and model optimization, we demonstrate highly accurate eye disease classification from fundus images, comparable to or superior to previous benchmarks. Full article

(This article belongs to the Special Issue Deep Learning Applications in Medical Imaging)

► Show Figures

Figure 1

31 pages, 18202 KiB

Open AccessArticle

MATLAB Application for Determination of 12 Combustion Products, Adiabatic Temperature and Laminar Burning Velocity: Development, Coding and Explanation

by Roberto Franco Cisneros and Freddy Jesus Rojas

Computation 2024, 12(9), 189; https://doi.org/10.3390/computation12090189 - 16 Sep 2024

Abstract

The determination of the characteristics and main combustion properties of fuels is necessary for post-implementation in different applications. Among the most important combustion properties of a fuel are the combustion products, flame temperature and laminar burning velocity. Therefore, this paper describes the step-by-step [...] Read more.

The determination of the characteristics and main combustion properties of fuels is necessary for post-implementation in different applications. Among the most important combustion properties of a fuel are the combustion products, flame temperature and laminar burning velocity. Therefore, this paper describes the step-by-step development and coding of a MATLAB application that can determine 12 combustion products, flame temperature and laminar burning velocity in order to understand the logic of calculus procedure, so any user would be able to make improvements of new functionalities (add more fuels, add more combustion products, etc.). The numerical procedure and methods (Gaussian elimination, Taylor Series and Newton–Raphson) parallel with their implementation as code lines for the development of the application are carried out using flow charts. In addition, simulations in Ansys Chemkin were performed and included in the application as part of the results comparison. It was found that: (1) The MATLAB Application codification and development were successfully explained in detail, (2) the functions and execution sequence are described by using flow charts and code extract, (3) the application is available to everyone for modifications, (4) the application can only be used for hydrocarbons fuels, (5) the application execution time registered was less than 8 s. Full article

► Show Figures

Figure 1

23 pages, 4202 KiB

Open AccessArticle

Performance Analysis and Optimization of a Channeled Photovoltaic Thermal System with Fin Absorbers and Combined Bi-Fluid Cooling

by Hamza Nasri, Jamel Riahi, Hatem Oueslati, Hichem Taghouti and Silvano Vergura

Computation 2024, 12(9), 188; https://doi.org/10.3390/computation12090188 - 15 Sep 2024

Abstract

The conversion efficiency of photovoltaic (PV) cells can be increased by reducing high temperatures with appropriate cooling. Passive cooling systems using air, water, ethylene glycol, and air/water+TiO₂ nano bi-fluid froth in the duct channel have been studied, but an overall assessment is [...] Read more.

The conversion efficiency of photovoltaic (PV) cells can be increased by reducing high temperatures with appropriate cooling. Passive cooling systems using air, water, ethylene glycol, and air/water+TiO₂ nano bi-fluid froth in the duct channel have been studied, but an overall assessment is essential for its possible application. In the present work, a numerical study is adopted to investigate the impact of the fluid-duct channel type on the electrical and thermal efficiency of the photovoltaic thermal (PVT) collector. Such investigation is achieved by means of a MATLAB R2022b code based on the Runge–Kutta (RK4) method. Four kinds of fluid duct channels are used to optimize the best fluid for improving the overall efficiency of the investigated PVT system. The numerical validation of the proposed model has been made by comparing the numerical and experimental results reported in the literature. The outcomes indicate that varying the duct channel nature affects mainly the electrical and thermal efficiency of the PVT collector. Our results validate that the nature of the fluid affects weakly the electrical efficiency, whereas the thermal efficiency is strongly affected. Accordingly, it is observed that PVT collectors based on nano bi-fluid air/water+TiO₂ give the best performance. In this context, an appreciable increase in the overall efficiency of 22% is observed when the water+TiO₂ fluid is substituted by air/ water+TiO₂ nano bi-fluid. Therefore, these motivating results make the PVT nano bi-fluid efficient and suitable for solar photovoltaic thermal applications since this system exhibits a daily overall efficiency of about 56.96%. The present work proves that controlling the design, cooling technique, and nature of the cooling fluid used is a crucial factor for improving the electrical, thermal, and overall efficiency of the PVT systems. Full article

(This article belongs to the Section Computational Engineering)

► Show Figures

Figure 1

14 pages, 4664 KiB

Open AccessArticle

Exploring Soliton Solutions for Fractional Nonlinear Evolution Equations: A Focus on Regularized Long Wave and Shallow Water Wave Models with Beta Derivative

by Sujoy Devnath, Maha M. Helmi and M. Ali Akbar

Computation 2024, 12(9), 187; https://doi.org/10.3390/computation12090187 - 11 Sep 2024

Abstract

The fractional regularized long wave equation and the fractional nonlinear shallow-water wave equation are the noteworthy models in the domains of fluid dynamics, ocean engineering, plasma physics, and microtubules in living cells. In this study, a reliable and efficient improved F-expansion technique, along [...] Read more.

The fractional regularized long wave equation and the fractional nonlinear shallow-water wave equation are the noteworthy models in the domains of fluid dynamics, ocean engineering, plasma physics, and microtubules in living cells. In this study, a reliable and efficient improved F-expansion technique, along with the fractional beta derivative, has been utilized to explore novel soliton solutions to the stated wave equations. Consequently, the study establishes a variety of reliable and novel soliton solutions involving trigonometric, hyperbolic, rational, and algebraic functions. By setting appropriate values for the parameters, we obtained peakons, anti-peakon, kink, bell, anti-bell, singular periodic, and flat kink solitons. The physical behavior of these solitons is demonstrated in detail through three-dimensional, two-dimensional, and contour representations. The impact of the fractional-order derivative on the wave profile is notable and is illustrated through two-dimensional graphs. It can be stated that the newly established solutions might be further useful for the aforementioned domains. Full article

(This article belongs to the Section Computational Engineering)

► Show Figures

Figure 1

21 pages, 1148 KiB

Open AccessArticle

Exploring Bifurcation in the Compartmental Mathematical Model of COVID-19 Transmission

by Olena Kiseleva, Sergiy Yakovlev, Dmytro Chumachenko and Oleksandr Kuzenkov

Computation 2024, 12(9), 186; https://doi.org/10.3390/computation12090186 - 11 Sep 2024

Abstract

This study proposes and theoretically substantiates a unique mathematical model for predicting the spread of infectious diseases using the example of COVID-19. The model is described by a special system of autonomous differential equations, which has scientific novelty for cases of complex dynamics [...] Read more.

This study proposes and theoretically substantiates a unique mathematical model for predicting the spread of infectious diseases using the example of COVID-19. The model is described by a special system of autonomous differential equations, which has scientific novelty for cases of complex dynamics of disease transmission. The adequacy of the model is confirmed by testing on the example of the spread of COVID-19 in one of the largest regions of Ukraine, both in terms of population and area. The practical novelty emerges through its versatile application in real-world contexts, guiding organizational decisions and public health responses. The model’s capacity to facilitate system functioning evaluation and identify significant parameters underlines its potential for proactive management and effective response in the evolving landscape of infectious diseases. Full article

► Show Figures

Figure 1

10 pages, 2363 KiB

Open AccessArticle

Computationally Efficient Outlier Detection for High-Dimensional Data Using the MDP Algorithm

by Michail Tsagris, Manos Papadakis, Abdulaziz Alenazi and Omar Alzeley

Computation 2024, 12(9), 185; https://doi.org/10.3390/computation12090185 - 11 Sep 2024

Abstract

Outlier detection, or anomaly detection as it is known in the machine learning community, has gained interest in recent years, and it is commonly used when the sample size is smaller than the number of variables. In 2015, an outlier detection procedure was [...] Read more.

Outlier detection, or anomaly detection as it is known in the machine learning community, has gained interest in recent years, and it is commonly used when the sample size is smaller than the number of variables. In 2015, an outlier detection procedure was proposed 7 for this high-dimensional setting, replacing the classic minimum covariance determinant estimator with the minimum diagonal product estimator. Computationally speaking, their method has two drawbacks: (a) it is not computationally efficient and does not scale up, and (b) it is not memory efficient and, in some cases, it is not possible to apply due to memory limits. We address the first issue via efficient code written in both R and C++, whereas for the second issue, we utilize the eigen decomposition and its properties. Experiments are conducted using simulated data to showcase the time improvement, while gene expression data are used to further examine some extra practicalities associated with the algorithm. The simulation studies yield a speed-up factor that ranges between 17 and 1800, implying a successful reduction in the estimator’s computational burden. Full article

(This article belongs to the Section Computational Biology)

► Show Figures

Figure 1

16 pages, 4761 KiB

Open AccessArticle

An Algorithm for Coding an Additive Manufacturing File from the Pressure Distribution of a Baropodometric Board for 3D Printing Customised Orthopaedic Insoles

by Francesco Simi, Gabriele Maria Fortunato, Fabio Diana, Jacopo Gai and Carmelo De Maria

Computation 2024, 12(9), 184; https://doi.org/10.3390/computation12090184 - 10 Sep 2024

Abstract

Customised orthotic insoles play a critical role in addressing foot pathologies and improving comfort and biomechanical alignment for patients with specific needs. The use of 3D printing technology for the manufacturing of orthotic insoles has received considerable attention in recent years due to [...] Read more.

Customised orthotic insoles play a critical role in addressing foot pathologies and improving comfort and biomechanical alignment for patients with specific needs. The use of 3D printing technology for the manufacturing of orthotic insoles has received considerable attention in recent years due to its potential for customisation, rapid prototyping, and cost-effectiveness. This paper presents the implementation of an algorithm purposely developed to generate an Additive Manufacturing File (AMF) containing the geometry of a patient-specific insole and the stiffness distribution based on pressure analysis from a baropodometric board. The generated file is used to 3D print via Fused Deposition Modelling an insole with a variable infill percentage depending on the pressure distribution on the patient’s foot. Three inputs are used as source data for the AMF file coding: (i) the 3D model that defines the geometry of the insole designed by the orthopaedist; (ii) the pressure map of the patient’s feet obtained with a baropodometric board; and (iii) the stiffness of the material that will be used to fabricate the insole. The proposed approach allows the fabrication of a patient-specific insole, capable of restoring the correct pressure distribution on the foot by varying the infill percentage. Two types of insoles were successfully fabricated using the implemented algorithm: the first was 3D printed, adding a top layer to be ready-to-use; the second was 3D printed without a top surface to be further customised with different coatings. The method described in this paper is robust for the fabrication of customised insoles and aims at overcoming the limitations of the traditional approach based on milling machining (e.g., time, costs, and path planning) since it can be easily integrated into any orthopaedic workshop. Full article

(This article belongs to the Section Computational Engineering)

► Show Figures

Figure 1

17 pages, 994 KiB

Open AccessArticle

Second-Order Modified Nonstandard Explicit Euler and Explicit Runge–Kutta Methods for n-Dimensional Autonomous Differential Equations

by Fawaz K. Alalhareth, Madhu Gupta, Hristo V. Kojouharov and Souvik Roy

Computation 2024, 12(9), 183; https://doi.org/10.3390/computation12090183 - 9 Sep 2024

Abstract

Nonstandard finite-difference (NSFD) methods, pioneered by R. E. Mickens, offer accurate and efficient solutions to various differential equation models in science and engineering. NSFD methods avoid numerical instabilities for large time steps, while numerically preserving important properties of exact solutions. However, most NSFD [...] Read more.

Nonstandard finite-difference (NSFD) methods, pioneered by R. E. Mickens, offer accurate and efficient solutions to various differential equation models in science and engineering. NSFD methods avoid numerical instabilities for large time steps, while numerically preserving important properties of exact solutions. However, most NSFD methods are only first-order accurate. This paper introduces two new classes of explicit second-order modified NSFD methods for solving n-dimensional autonomous dynamical systems. These explicit methods extend previous work by incorporating novel denominator functions to ensure both elementary stability and second-order accuracy. This paper also provides a detailed mathematical analysis and validates the methods through numerical simulations on various biological systems. Full article

(This article belongs to the Special Issue Advanced Numerical Methods for Solving Differential Equations with Applications in Science and Engineering)

► Show Figures

Figure 1

10 pages, 2344 KiB

Open AccessArticle

An Analysis of the Stress–Strain State of a Layer on Two Cylindrical Bearings

by Vitaly Miroshnikov, Oleksandr Denshchykov, Iaroslav Grebeniuk and Oleksandr Savin

Computation 2024, 12(9), 182; https://doi.org/10.3390/computation12090182 - 6 Sep 2024

Abstract

A spatial problem of elasticity theory is solved for a layer located on two bearings embedded in it. The bearings are represented as thick-walled pipes embedded in the layer parallel to its boundaries. The pipes are rigidly connected to the layer, and contact-type [...] Read more.

A spatial problem of elasticity theory is solved for a layer located on two bearings embedded in it. The bearings are represented as thick-walled pipes embedded in the layer parallel to its boundaries. The pipes are rigidly connected to the layer, and contact-type conditions (normal displacements and tangential stresses) are specified on the insides of the pipes. Stresses are set on the flat surfaces of the layer. The objective of this study is to obtain the stress–strain state of the body of the layer under different geometric characteristics of the model. The solution to the problem is presented in the form of the Lamé equation, whose terms are written in different coordinate systems. The generalized Fourier method is used to transfer the basic solutions between coordinate systems. By satisfying the boundary and conjugation conditions, the problem is reduced to a system of infinite linear algebraic equations of the second kind, to which the reduction method is applied. After finding the unknowns, using the generalized Fourier method, it is possible to find the stress–strain state at any point of the body. The numerical study of the stress state showed high convergence of the approximate solutions to the exact one. The stress–strain state of the composite body was analyzed for different geometric parameters and different pipe materials. The results obtained can be used for the preliminary determination of the geometric parameters of the model and the materials of the joints. The proposed solution method can be used not only to calculate the stress state of bearing joints, but also of bushings (under specified conditions of rigid contact without friction on the internal surfaces). Full article

(This article belongs to the Special Issue Integrated Computer Technologies in Mechanical Engineering—Synergetic Engineering III)

► Show Figures

Figure 1

29 pages, 16569 KiB

Open AccessArticle

Some Aspects of the Effects of Dry Friction Discontinuities on the Behaviour of Dynamic Systems

by Stelian Alaci, Costica Lupascu, Ionut-Cristian Romanu, Delia-Aurora Cerlinca and Florina-Carmen Ciornei

Computation 2024, 12(9), 181; https://doi.org/10.3390/computation12090181 - 5 Sep 2024

Abstract

Most studies in the literature consider the value of the coefficient of dynamic friction to be constant. We studied the evolution of a dynamic system when the coefficient of friction results in different values depending on the contact surfaces. A system with four [...] Read more.

Most studies in the literature consider the value of the coefficient of dynamic friction to be constant. We studied the evolution of a dynamic system when the coefficient of friction results in different values depending on the contact surfaces. A system with four balls fixed on an aluminium plate was driven with constant velocity into motion on the coaxial races of two identical outer bearing rings. The assembly presents a motion with periodic variable amplitude between two extremes, a fact that suggests the presence of a periodical excitation. The test was repeated, but this time, new bodies were used, which were two identical bodies made of two balls rigidised via a short cylindrical rod. When the rings were driven into rotational motion, the two bodies performed different motions; if the bodies were inter-changed, the differences between the motions remained. The rings were analysed, and a small region on the race of one ring was observed, where the roughness was considerably greater than the rest of the surface. Then, a mathematical model for the dynamic system with different friction coefficients was proposed and solved. This model is capable of simulating different situations, such as oscillatory motion and circular motion, with or without separation of the contacting bodies. Here, we present a dynamic model with Hertzian contact points in the presence of dry friction, with the coefficient of friction changing suddenly on the contacting surfaces. Full article

(This article belongs to the Section Computational Engineering)

► Show Figures

Figure 1

20 pages, 1018 KiB

Open AccessArticle

Optimized Calculation of Radial and Axial Magnetic Forces between Two Non-Coaxial Coils of Rectangular Cross-Section with Parallel Axes

by Slobodan Babic, Eray Guven, Kai-Hong Song and Yao Luo

Computation 2024, 12(9), 180; https://doi.org/10.3390/computation12090180 - 4 Sep 2024

Abstract

In this paper, we introduce a new algorithm for calculating the radial and axial magnetic forces between two non-coaxial circular loops with parallel axes. These formulas are derived from a modified version of Grover’s formula for mutual inductance between the coils in question. [...] Read more.

In this paper, we introduce a new algorithm for calculating the radial and axial magnetic forces between two non-coaxial circular loops with parallel axes. These formulas are derived from a modified version of Grover’s formula for mutual inductance between the coils in question. Utilizing these formulas, we compute the radial and axial magnetic forces between two non-coaxial thick coils of rectangular cross-sections with parallel axes. In these calculations, we apply the filament method and conduct investigations to determine the optimal number of subdivisions for the coils in terms of computational time and accuracy. The method presented in this paper is also applicable to all conventional non-coaxial coils, such as disks, solenoids, and non-conventional coils like Bitter coils, all with parallel axes. This paper emphasizes the accuracy and computational efficiency of the calculations. Furthermore, the new method is validated according to several previously established methods. Full article

(This article belongs to the Section Computational Engineering)

► Show Figures

Figure 1

21 pages, 2205 KiB

Open AccessArticle

PyIncentiveBC: A Python Module for Simulation of Incentivization Mechanism Implemented in Blockchain-Based Systems

by Abdellah Ouaguid, Mohamed Hanine, Zouhair Chiba, Noreddine Abghour and Mohammed Ouzzif

Computation 2024, 12(9), 179; https://doi.org/10.3390/computation12090179 - 3 Sep 2024

Abstract

The diversity of approaches for retaining participants in a Blockchain-based system complicates benchmarking. The majority of proposals for rewarding and penalizing participants in these systems are limited to their own set of data and scenarios, making it hard to compare their effectiveness. To [...] Read more.

The diversity of approaches for retaining participants in a Blockchain-based system complicates benchmarking. The majority of proposals for rewarding and penalizing participants in these systems are limited to their own set of data and scenarios, making it hard to compare their effectiveness. To overcome these challenges, we developed PyIncentiveBC, a free, open-source, and modular simulator designed to evaluate the reliability of any approach, incorporating a dynamic and proportionate incentivization mechanism proposed in our previous work. PyIncentiveBC aims to provide the scientific communities with an extensible software solution facilitating the benchmarking of existing approaches with new ones proposed by them. Full article

(This article belongs to the Section Computational Engineering)

► Show Figures

Figure 1

16 pages, 12492 KiB

Open AccessArticle

The Development of a Novel Transient Signal Analysis: A Wavelet Transform Approach

by Eduardo Gómez-Luna, Dixon E. Cuadros-Orta, John E. Candelo-Becerra and Juan C. Vasquez

Computation 2024, 12(9), 178; https://doi.org/10.3390/computation12090178 - 3 Sep 2024

Abstract

This paper presents a new method for the analysis of transient signals in the frequency domain based on the Continuous Wavelet Transform (CWT). The proposed case study involves test signals measured from an electronic switch considering open and close operations. The source is [...] Read more.

This paper presents a new method for the analysis of transient signals in the frequency domain based on the Continuous Wavelet Transform (CWT). The proposed case study involves test signals measured from an electronic switch considering open and close operations. The source is connected to inductive, resistive, and capacitive loads. Resonance behaviors are introduced and compared with the Discrete Fourier Transform (DFT). Multiple factors, such as reliability, repeatability, high noise attenuation, and the smoothing of the analyzed spectrum, are considered in this study. This proposed study highlights the effectiveness of CWT in signal processing, especially in obtaining a detailed spectrum that reveals the behavior of electrical circuits. Resonance behaviors were analyzed, demonstrating that the signal processing performed by CWT is better for spectrum analysis than DFT. This study shows the potential of CWT to analyze transient electrical signals, specifically for identifying and characterizing the behavior of load connections and disconnections. Full article

(This article belongs to the Section Computational Engineering)

► Show Figures

Figure 1

3 pages, 315 KiB

Open AccessCorrection

Correction: Oldrieve, R.M. Teaching K–3 Multi-Digit Arithmetic Computation to Students with Slow Language Processing. Computation 2024, 12, 128

by Richard M. Oldrieve

Computation 2024, 12(9), 177; https://doi.org/10.3390/computation12090177 - 3 Sep 2024

Abstract

Error in Figure/Table [...] Full article

(This article belongs to the Special Issue Computations in Mathematics, Mathematical Education, and Science)

► Show Figures

Figure 10

17 pages, 4355 KiB

Open AccessArticle

On the Impact of Discrete Atomic Compression on Image Classification by Convolutional Neural Networks

by Viktor Makarichev, Vladimir Lukin and Iryna Brysina

Computation 2024, 12(9), 176; https://doi.org/10.3390/computation12090176 - 1 Sep 2024

Abstract

Digital images play a particular role in a wide range of systems. Image processing, storing and transferring via networks require a lot of memory, time and traffic. Also, appropriate protection is required in the case of confidential data. Discrete atomic compression (DAC) is [...] Read more.

Digital images play a particular role in a wide range of systems. Image processing, storing and transferring via networks require a lot of memory, time and traffic. Also, appropriate protection is required in the case of confidential data. Discrete atomic compression (DAC) is an approach providing image compression and encryption simultaneously. It has two processing modes: lossless and lossy. The latter one ensures a higher compression ratio in combination with inevitable quality loss that may affect decompressed image analysis, in particular, classification. In this paper, we explore the impact of distortions produced by DAC on performance of several state-of-the-art classifiers based on convolutional neural networks (CNNs). The classic, block-splitting and chroma subsampling modes of DAC are considered. It is shown that each of them produces a quite small effect on MobileNetV2, VGG16, VGG19, ResNet50, NASNetMobile and NASNetLarge models. This research shows that, using the DAC approach, memory expenses can be reduced without significant degradation of performance of the aforementioned CNN-based classifiers. Full article

(This article belongs to the Special Issue Integrated Computer Technologies in Mechanical Engineering—Synergetic Engineering III)

► Show Figures

Figure 1

29 pages, 10764 KiB

Open AccessArticle

In Silico Drug Screening for Hepatitis C Virus Using QSAR-ML and Molecular Docking with Rho-Associated Protein Kinase 1 (ROCK1) Inhibitors

by Joshua R. De Borja and Heherson S. Cabrera

Computation 2024, 12(9), 175; https://doi.org/10.3390/computation12090175 - 31 Aug 2024

Abstract

The enzyme ROCK1 plays a pivotal role in the disruption of the tight junction protein CLDN1, a downstream effector influencing various cellular functions such as cell migration, adhesion, and polarity. Elevated levels of ROCK1 pose challenges in HCV, where CLDN1 serves as a [...] Read more.

The enzyme ROCK1 plays a pivotal role in the disruption of the tight junction protein CLDN1, a downstream effector influencing various cellular functions such as cell migration, adhesion, and polarity. Elevated levels of ROCK1 pose challenges in HCV, where CLDN1 serves as a crucial entry factor for viral infections. This study integrates a drug screening protocol, employing a combination of quantitative structure–activity relationship machine learning (QSAR-ML) techniques; absorption, distribution, metabolism, and excretion (ADME) predictions; and molecular docking. This integrated approach allows for the effective screening of specific compounds, using their calculated features and properties as guidelines for selecting drug-like candidates targeting ROCK1 inhibition in HCV treatment. The QSAR-ML model, validated with scores of 0.54 (R²), 0.15 (RMSE), and 0.71 (CCC), demonstrates its predictive capabilities. The ADME-Docking study’s final results highlight notable compounds from ZINC15, specifically ZINC000071318464, ZINC000073170040, ZINC000058568630, ZINC000058591055, and ZINC000058574949. These compounds exhibit the best ranking Vina scores for protein–ligand binding with the crystal structure of ROCK1 at the C2 pocket site. The generated features and calculated pIC₅₀ bioactivity of these compounds provide valuable insights, facilitating the identification of structurally similar candidates in the ongoing exploration of drugs for ROCK1 inhibition. Full article

(This article belongs to the Special Issue 10th Anniversary of Computation—Computational Biology)

► Show Figures

Figure 1

18 pages, 6919 KiB

Open AccessArticle

FPGA-Based Numerical Simulation of the Chaotic Synchronization of Chua Circuits

by Leonardo Rentería, Margarita Mayacela, Klever Torres, Wladimir Ramírez, Rolando Donoso and Rodrigo Acosta

Computation 2024, 12(9), 174; https://doi.org/10.3390/computation12090174 - 31 Aug 2024

Abstract

The objective of this work was to design and implement a system based on reconfigurable hardware as a study tool for the synchronization of chaotic circuits. Mathematical models were established for one circuit, two synchronized, and multiple synchronized Chua circuits. An ordinary differential [...] Read more.

The objective of this work was to design and implement a system based on reconfigurable hardware as a study tool for the synchronization of chaotic circuits. Mathematical models were established for one circuit, two synchronized, and multiple synchronized Chua circuits. An ordinary differential equation solver was developed applying Euler’s method using the Verilog hardware description language and synthesized on a Spartan 3E FPGA (Field-Programmable Gate Array) equipped with a 32-bit RISC processor, 64 MB of DDR SDRAM, and 4 Mb of PROM. With a step size of 0.005 and a total of 10,000 iterations, the state equations for one and three Chua circuits were solved at a time of 0.2 ms and a frequency of 50 Mhz. The logical resources used by the system did not exceed 4%. To verify the operation, a numerical simulation was carried out using the Octave V9.1.0 calculation software on an Intel(R) Core i7-9750H CPU 2.59 GHz computer, obtaining the same results but in a time of 493 ms and 3.177 s for one and three circuits, respectively. Full article

(This article belongs to the Section Computational Engineering)

► Show Figures

Figure 1

22 pages, 13050 KiB

Open AccessArticle

A Deep Learning Model for Detecting Fake Medical Images to Mitigate Financial Insurance Fraud

by Muhammad Asad Arshed, Shahzad Mumtaz, Ștefan Cristian Gherghina, Neelam Urooj, Saeed Ahmed and Christine Dewi

Computation 2024, 12(9), 173; https://doi.org/10.3390/computation12090173 - 29 Aug 2024

Abstract

Artificial Intelligence and Deepfake Technologies have brought a new dimension to the generation of fake data, making it easier and faster than ever before—this fake data could include text, images, sounds, videos, etc. This has brought new challenges that require the faster development [...] Read more.

Artificial Intelligence and Deepfake Technologies have brought a new dimension to the generation of fake data, making it easier and faster than ever before—this fake data could include text, images, sounds, videos, etc. This has brought new challenges that require the faster development of tools and techniques to avoid fraudulent activities at pace and scale. Our focus in this research study is to empirically evaluate the use and effectiveness of deep learning models such as Convolutional Neural Networks (CNNs) and Patch-based Neural Networks in the context of successful identification of real and fake images. We chose the healthcare domain as a potential case study where the fake medical data generation approach could be used to make false insurance claims. For this purpose, we obtained publicly available skin cancer data and used recently introduced stable diffusion approaches—a more effective technique than prior approaches such as Generative Adversarial Network (GAN)—to generate fake skin cancer images. To the best of our knowledge, and based on the literature review, this is one of the few research studies that uses images generated using stable diffusion along with real image data. As part of the exploratory analysis, we analyzed histograms of fake and real images using individual color channels and averaged across training and testing datasets. The histogram analysis demonstrated a clear change by shifting the mean and overall distribution of both real and fake images (more prominent in blue and green) in the training data whereas, in the test data, both means were different from the training data, so it appears to be non-trivial to set a threshold which could give better predictive capability. We also conducted a user study to observe where the naked eye could identify any patterns for classifying real and fake images, and the accuracy of the test data was observed to be 68%. The adoption of deep learning predictive approaches (i.e., patch-based and CNN-based) has demonstrated similar accuracy (~100%) in training and validation subsets of the data, and the same was observed for the test subset with and without StratifiedKFold (k = 3). Our analysis has demonstrated that state-of-the-art exploratory and deep-learning approaches are effective enough to detect images generated from stable diffusion vs. real images. Full article

(This article belongs to the Special Issue Computational Medical Image Analysis—2nd Edition)

► Show Figures

Figure 1

Journal Description

Computation

Latest Articles

Journal Menu

Journal Browser

Highly Accessed Articles

Latest Books

E-Mail Alert

News

Topics

Conferences

Special Issues

Further Information

Guidelines

MDPI Initiatives

Follow MDPI