Showing papers in "European Physical Journal-special Topics in 2022"

Artificial neural networks: a practical review of applications involving fractional calculus

Abstract: In this work, a bibliographic analysis on artificial neural networks (ANNs) using fractional calculus (FC) theory has been developed to summarize the main features and applications of the ANNs. ANN is a mathematical modeling tool used in several sciences and engineering fields. FC has been mainly applied on ANNs with three different objectives, such as systems stabilization, systems synchronization, and parameters training, using optimization algorithms. FC and some control strategies have been satisfactorily employed to attain the synchronization and stabilization of ANNs. To show this fact, in this manuscript are summarized, the architecture of the systems, the control strategies, and the fractional derivatives used in each research work, also, the achieved goals are presented. Regarding the parameters training using optimization algorithms issue, in this manuscript, the systems types, the fractional derivatives involved, and the optimization algorithm employed to train the ANN parameters are also presented. In most of the works found in the literature where ANNs and FC are involved, the authors focused on controlling the systems using synchronization and stabilization. Furthermore, recent applications of ANNs with FC in several fields such as medicine, cryptographic, image processing, robotic are reviewed in detail in this manuscript. Works with applications, such as chaos analysis, functions approximation, heat transfer process, periodicity, and dissipativity, also were included. Almost to the end of the paper, several future research topics arising on ANNs involved with FC are recommended to the researchers community. From the bibliographic review, we concluded that the Caputo derivative is the most utilized derivative for solving problems with ANNs because its initial values take the same form as the differential equations of integer-order.

COVID-19 disease diagnosis with light-weight CNN using modified MFCC and enhanced GFCC from human respiratory sounds

Abstract: In the last 2 years, medical researchers and clinical scientists have paid close attention to the problem of respiratory sound classification to classify COVID-19 disease symptoms. In the physical world, very few AI-based (Artificial Intelligence) techniques are often used to detect COVID-19/SARS-CoV-2 respiratory disease symptoms from the human respiratory system-generated acoustic sounds such as acoustic voice sound, breathing (inhale and exhale) sounds, and cough sound. We propose a light-weight Convolutional Neural Network (CNN) with Modified-Mel-frequency Cepstral Coefficient (M-MFCC) using different depths and kernel sizes to classify COVID-19 and other respiratory sound disease symptoms such as Asthma, Pertussis, and Bronchitis. The proposed network outperforms conventional feature extraction models and existing Deep Learning (DL) models for COVID-19/SARS-CoV-2 classification accuracy in the range of 4–10%. The model’s performance is compared with the COVID-19 crowdsourced benchmark dataset and gives a competitive performance. We applied different receptive fields and depths in the proposed model to get different contextual information that should aid in classification. And our experiments suggested 1 $$\times $$ 12 receptive fields and a depth of 5-Layer for the light-weight CNN to extract and identify the features from respiratory sound data. The model is also trained and tested with different modalities of data to showcase its effectiveness in classification.

Estimating a continuously varying offset between multivariate time series with application to COVID-19 in the United States

Abstract: This paper introduces new methods to track the offset between two multivariate time series on a continuous basis. We then apply this framework to COVID-19 counts on a state-by-state basis in the United States to determine the progression from cases to deaths as a function of time. Across multiple approaches, we reveal an "up-down-up" pattern in the estimated offset between reported cases and deaths as the pandemic progresses. This analysis could be used to predict imminent increased load on a healthcare system and aid the allocation of additional resources in advance.

A mathematical model for SARS-CoV-2 in variable-order fractional derivative

Abstract: A new coronavirus mathematical with hospitalization is considered with the consideration of the real cases from March 06, 2021 till the end of April 30, 2021. The essential mathematical results for the model are presented. We show the model stability when $${\mathcal {R}}_0<1$$ in the absence of infection. We show that the system is stable locally asymptotically when $${\mathcal {R}}_0<1$$ at infection free state. We also show that the system is globally asymptotically stable in the disease absence when $${\mathcal {R}}_0<1$$ . Data have been used to fit accurately to the model and found the estimated basic reproduction number to be $$ {\mathcal {R}}_0= 1.2036$$ . Some graphical results for the effective parameters are drawn for the disease elimination. In addition, a variable-order model is introduced, and so as to handle the outbreak effectively and efficiently, a genetic algorithm is used to produce high-quality control. Numerical simulations clearly show that decision-makers may develop helpful and practical strategies to manage future waves by implementing optimum policies.

Fake or real news about COVID-19? Pretrained transformer model to detect potential misleading news

Abstract: The World Health Organization declared the novel coronavirus disease 2019 a pandemic on March 11, 2020. Along with the coronavirus pandemic, a new crisis has emerged, characterized by widespread fear and panic caused by a lack of information or, in some cases, outright fake messages. In these circumstances, Twitter is one of the most eminent and trusted social media platforms. Fake tweets, on the other hand, are challenging to detect and differentiate. The primary goal of this paper is to educate society about the importance of accurate information and prevent the spread of fake information. This paper has investigated COVID-19 fake data from various social media platforms such as Twitter, Facebook, and Instagram. The objective of this paper is to categorize given tweets as either fake or real news. The authors have tested various deep learning models on the COVID-19 fake dataset. Finally, the CT-BERT and RoBERTa deep learning models outperformed other deep learning models like BERT, BERTweet, AlBERT, and DistlBERT. The proposed ensemble deep learning architecture outperformed CT-BERT and RoBERTa on the COVID-19 fake news dataset using the multiplicative fusion technique. The proposed model's performance in this technique was determined by the multiplicative product of the final predictive values of CT-BERT and RoBERTa. This technique overcomes the disadvantage of these CT-BERT and RoBERTa models' incorrect predictive nature. The proposed architecture outperforms both well-known ML and DL models, with 98.88% accuracy and a 98.93% F1-score.

Hybrid nanofluid flow around a triangular-shaped obstacle inside a split lid-driven trapezoidal cavity

Abstract: Numerical simulation analyzes the mixed convection flow of $$\hbox {Al}_{2}\hbox {O}_{3}{-}\hbox {Cu}{-}\hbox {H}_{2}\hbox {O}$$ (aluminium oxide–copper–water) hybrid nanofluid inside a split lid-driven trapezoidal cavity. A triangular-shaped cold obstacle is placed inside the cavity. The horizontal base of the cavity is kept cold, whereas the side walls are chosen adiabatic. The thermally active upper wall maintained at a constant temperature is split into halves, and each half moves opposite to the other with constant velocity. Modeled equations are converted into a nonlinear system of partial differential equations. This system, along with incorporated physical boundary constraints, is solved numerically via Galerkin finite-element method. Attained results are also compared with the earlier publications to ensure validation and accuracy. To examine the effects of various pertinent parameters, various flow and heat transfer attributes like dimensionless velocity, stream contours, temperature, and isotherms, and local and average Nusselt numbers are critically analyzed. The outcomes of this examination will provide qualitative suggestions to improve the cooling mechanism of several electronic gadgets and thermal devices.

High-speed 3D imaging using a chaos lidar system

Abstract: Abstract We characterize a new chaos lidar system configuration and demonstrate its capability for high-speed 3D imaging. Compared with a homodyned scheme employing single-element avalanche photodetectors (APDs), the proposed scheme utilizes a fiber Bragg grating and quadrant APDs to substantially increase the system throughput, frame rate, and field-of-view. By quantitatively analyzing the signal-to-noise ratio, peak-to-standard deviation of the sidelobe level, precision, and detection probability, we show that the proposed scheme has better detection performance suitable for practical applications. To show the feasibility of the chaos lidar system, while under the constrain of eye-safe regulation, we demonstrate high-speed 3D imaging with indoor and outdoor scenes at a throughput of 100 kHz, a frame rate of 10 Hz, and a FOV of 24.5 $$^\circ $$ ∘ $$\times $$ × 11.5 $$^\circ $$ ∘ for the first time.

Novel compressive sensing image encryption using the dynamics of an adjustable gradient Hopfield neural network

Abstract: Abstract In this contribution, the nonlinear dynamics of a non-autonomous model of two neurons based on the Hopfield neural network is considered. Using activation gradients as bifurcation control parameters, the properties of the model include dissipation with the existence of attractors and equilibrium points with their stability. Using traditional nonlinear analysis tools such as bifurcation diagrams, the graph of the maximum Lyapunov exponent, phase portraits, two-parameter diagrams, and attraction basins, the complex behaviour of the two-dimensional Hopfield neural network has been investigated and several windows of multistability involving the coexistence of up to four coexisting attractors have been found. Besides, the results of our numerical simulation of the multistability have been further supported using some Pspice simulation. The effect of the fractional-order derivative is also explored, and it is found that the route toward chaos is completely different when the order q of the HNN is varied between $$0<q<1$$ 0 < q < 1 . Finally, a compressive sensing approach is used to compress and encrypt color images based on the sequences of the above-mentioned system. The plain color image is decomposed into Red, Green, and Blue components. The Discrete Wavelet Transform (DWT) is applied to each component to obtain the corresponding sparse components. Confusion keys are obtained from the proposed chaotic system to scramble each sparse component. The measurement matrices obtained from the chaotic sequence are used to compress the confused sparse matrices corresponding to the Red, Green, and Blue components. Each component is quantified and a diffusion step is then applied to improve the randomness and, consequently, the information entropy. Experimental analysis of the proposed method yields a running time (t) of 6.85 ms, a maximum entropy value of 7.9996 for global and 7.9153 for local, an encryption throughput (ET) value of 114.80, and a number of cycles (NC) of 20.90. Analysis of these metrics indicates that the proposed scheme is competitive with some recent literature.

The modeling and analysis of the COVID-19 pandemic with vaccination and treatment control: a case study of Maharashtra, Delhi, Uttarakhand, Sikkim, and Russia in the light of pharmaceutical and non-pharmaceutical approaches

Abstract: Nonlinear dynamics is an exciting approach to describe the dynamical practices of COVID-19 disease. Mathematical modeling is a necessary method for investigating the dynamics of epidemic diseases. In the current article, an effort has been made to cultivate a novel COVID-19 compartment mathematical model by incorporating vaccinated populations. Primarily, the fundamental characteristics of the model, such as positivity and boundedness of solutions, are established. Thereafter, equilibrium analysis of steady states has been illustrated through vaccine reproduction number. Further, a nonlinear least square curve fitting technique has been employed to recognize the best fitted model parameters from the COVID-19 mortality data of five regions, namely Maharashtra, Delhi, Uttarakhand, Sikkim, and Russia. The numerical framework of the model has been added to interpret the consequence of various control schemes (pharmaceutical or non-pharmaceutical) on COVID-19 dynamics, and it has been ascertained that all the control protocols have a positive influence on curtailing the COVID-19 transference in the aforementioned regions. In addition, the essence of vaccine efficacy and vaccine-induced immunity are examined by considering different scenarios. Our analysis demonstrates that the disease will be wiped off from the Maharashtra, Delhi, Uttarakhand and Sikkim regions of India, while it shall persist in Russia for some more time. It is also found that, if a vaccine calamity arises, the government should majorly focus on permanent drug treatment of hospitalized individuals rather than vaccination.