The COVID-19 disease is one of the most severe health disasters encountered in the world recently. COVID-19 disease is a highly fatal disease that spreads quickly through the respiratory tract. The most accurate method for diagnosing COVID-19 disease is Reverse Transcription Polymerase Chain Reaction (RT-PCR). Although this method gives more precise results, it is extremely costly in terms of cost and processing time. To overcome this issue, many studies are being carried out, particularly on diagnosis via lung images. Deep learning, artificial intelligence, and machine learning methods are applied in image enhancement studies, and powerful results are obtained. Within the scope of this study, texture analysis methods, which are part of the image enhancement methods, are examined. Among these methods, the local binary pattern (LBP) method has a special place. Due to the extremely high success rate of the LBP method, many successful derivative methods have been derived from this method. The paper includes the developed quadstar local binary pattern (QS-LBP) method derived from the standard LBP method. Moreover, machine learning (ML) algorithms are used to process the features extracted from two different models using the QS-LBP method. Extremely high analysis values have been obtained from the hybrid method consisting of the developed models and ML methods, accuracy is 99.21%, kappa is 96.2%, ROC area is 99.8%, F-measure is 99.6 %, precision is 99.5%, and finally recall is 99.8%, respectively. Using the hybrid technique suggested in the paper, diagnosing COVID-19 disease in lung images has been carried out exceedingly successfully.

Differential evolution (DE) has demonstrated its significant contribution to the optimization of different real-world applications as well as standard benchmarks. This paper presents a novel variant of the DE algorithm, known as the LSHADESPA algorithm. The LSHADESPA algorithm incorporates three significant modifications to enhance its performance. Firstly, a proportional shrinking population mechanism is employed to reduce the computational burden. Secondly, a simulated annealing (SA)-based scaling factor is introduced to improve the exploration properties of the algorithm. Finally, an oscillating inertia weight-based crossover rate is utilized to strike a balance between exploitation and exploration. These modifications aim to enhance the overall efficiency and effectiveness of the DE algorithm. The proposed LSHADESPA algorithm has been empirically evaluated on a set of benchmark problems, namely the CEC 2014, CEC 2017, CEC 2021, as well as CEC 2022. The experimental outcomes show that the LSHADESPA algorithm performs superior to other MH algorithms. Additionally, the Wilcoxon rank-sum as well as the Friedman rank test proves the statistical significance of the proposed LSHADESPA in comparison to other algorithms under comparison. The outcomes indicate that the LSHADESPA algorithm has statistical significance, with Friedman statistics for the CEC 2014, CEC 2017, and CEC 2022 benchmark functions achieving the lowest f-rank value compared to the other MH algorithms which are found to be 41, 77, and 26, respectively, and obtained 1st rank. Note that this paper is an invited extended version of the paper published in ISCMI 2022 conference.

Abstract The early detection of structural damage is of paramount importance for the maintenance and prolongation of the service life of existing bridges. The identification of structural damage is typically conducted using nondestructive vibration experiments in conjunction with a mathematical technique known as model updating. This paper presents the application of two distinct model updating methodologies. The first method employs a metaheuristic optimization technique with nested multilevel sampling to efficiently explore the design parameter space. The second method frames model updating as an inverse problem and employs a machine learning-based surrogate model to approximate the inverse relationship between the structural response and the structural parameters. Based on the analysis of both approaches, a new method is proposed that combines the advantages of both methods in a novel way. The method employs iterative targeting and design space reduction, in conjunction with the use of an artificial neural network, to approximate the inverse design space as accurately as possible. This is achieved by sequential training the network on the optimal data set. This procedure significantly accelerates the convergence of the optimization process and improves the accuracy of the obtained solution. All methods are applied to the damage identification of three structures: single-span and two-span steel plate girders and a riveted truss bridge. The ability to detect damage in varying locations and over different scales is assessed. The results obtained are compared in terms of the convergence speed and the accuracy of the solutions obtained. Additionally, different settings of methods varying in the number of simulations, levels and the parameter controlling the convergence rate are also studied. In all the tested cases, the superiority of the newly proposed method in terms of accuracy and computational requirements has been confirmed. The new method, while maintaining the same number of simulations, on average increases the accuracy of the obtained results by a factor of 30. It also identifies damage with equal or better accuracy at one-tenth the number of simulations than other tested methods.