Search

Background: Moth-flame optimization will meet the premature and stagnation phenomenon when encountering difficult optimization tasks. Objective: To overcome the above shortcomings, this paper presented a quasi-reflection moth-flame optimization algorithm with refraction learning called QRMFO to strengthen the property of ordinary MFO and apply it in various application fields. Method: In the proposed QRMFO, quasi-reflection-based learning increases the diversity of the population and expands the search space on the iteration jump phase; refraction learning improves the accuracy of the potential optimal solution. Results: Several experiments are conducted to evaluate the superiority of the proposed QRMFO in the paper; first of all, the CEC2017 benchmark suite is utilized to estimate the capability of QRMFO when dealing with the standard test sets compared with the state-of-the-art algorithms; afterward, QRMFO is adopted to deal with multilevel thresholding image segmentation problems and real medical diagnosis case. Conclusion: Simulation results and discussions show that the proposed optimizer is superior to the basic MFO and other advanced methods in terms of convergence rate and solution accuracy.

Harris hawks optimization (HHO) has been accepted as one of the well-established swarm-based methods in the community of optimization and machine learning that primarily works based on multiple dynamic features and various exploratory and exploitative traits. Compared with other optimization algorithms, it has been observed that HHO can obtain high-quality solutions for continuous and constrained complex and real-world problems. While there is a wide variety of strategies in the HHO for dealing with diverse situations, there are chances for sluggish performance, where the convergence rate can gradually slow with time, and the HHO may stay stuck in the current relatively better place and may be unable to explore other better areas. To mitigate this concern, this paper combines the Cauchy mutation mechanism into the HHO algorithm named CMHHO. This idea can boost performance and provide a promising optimizer for solving complex optimization problems. The Cauchy mutation mechanism can speed up the convergence of the solution and help HHO explore more promising regions compared to its basic release. On 30 IEEE CEC2017 benchmark functions, the study compared the proposed CMHHO with various conventional and advanced metaheuristics to validate its performance and quality of solutions. It has been found through experiments that the overall optimization performance of CMHHO is far superior to all competitors. The CMHHO method is applied to four engineering challenges to investigate the capabilities of the proposed algorithm in solving real-world problems, and experimental results show that the suggested algorithm is more successful than existing algorithms.

Harris hawks optimization has been a popular swarm intelligence algorithm in recent years. In order to improve the local exploitation ability of the algorithm and improve the problem of slow convergence speed, an enhanced Harris hawks optimization algorithm based on Laplace crossover and random replacement strategy is proposed. This variant uses two optimization mechanisms. Firstly, Laplace crossover is added to enhance the exploitation ability of the algorithm. At the same time, the random replacement strategy is introduced into the original algorithm, which accelerates the convergence speed. The basic functions, IEEE CEC2011 and IEEE CEC2017 functions are used for algorithms comparison, balance diversity analysis, and high-dimensional experiments to verify the superiority of the algorithm proposed in this paper. The experimental results show that the improved algorithm has the advantages of strong optimization ability, high convergence accuracy, and fast convergence speed. The algorithm has solved five engineering design problems using these advantages and can effectively deal with constraint problems.

The whale optimizer is a popular metaheuristic algorithm, which has the problems of weak global exploration, easy falling into local optimum, and low optimization accuracy when searching for the optimal solution. To solve these problems, this paper proposes an enhanced whale optimization algorithm (WOA) based on the worst individual disturbance (WD) and neighborhood mutation search (NM), named WDNMWOA, which employed WD to enhance the ability to jump out of local optimum and global exploration, adopted NM to enhance the possibility of individuals approaching the optimal solution. The superiority of WDNMWOA is demonstrated by representative IEEE CEC2014, CEC2017, CEC2019, and CEC2020 benchmark functions and four engineering examples. The experimental results show that thes WDNMWOA has better convergence accuracy and strong optimization ability than the original WOA.

This study proposed an improved dragonfly algorithm (DA). This algorithm is a recently proposed metaheuristic optimizer inspired by swarming behaviors of dragonflies, which has reasonably achieved satisfactory results in dealing with engineering, education, and other fields. However, the original method will show some shortcomings in convergence speed or falling into local optimum. Given these shortcomings, this paper proposes an improved optimizer to balance the relationship between exploitation and exploration and mitigate any deficiency. First, by implementing the idea of the quantum rotation gate, the swarm of agents can be shifted to a position more conducive to the optimal value. Then, Gaussian mutation is adopted to improve the swarm’s ability to mutate and realize its diversity, which enables the primary method to have a strong local search capability. The proposed method was compared against six other common meta-heuristics and five state-of-the-art algorithms on a comprehensive set of nineteen functions selected from twenty-three classic benchmark problems and thirty IEEE (Institute of Electrical and Electronics Engineers) CEC (Congress on Evolutionary Computation) 2014 benchmark tasks. To verify the effectiveness of the approach, the non-parametric statistical Wilcoxon signed-rank and Friedman tests were performed to validate the significance of the proposed method against other counterparts. The results of experimental simulations demonstrate that two introduced strategies can significantly improve the exploitative and exploratory tendencies of the original algorithm. Furthermore, the convergence speed of the conventional approach has been improved to a large extent. Additionally, quantum-behaved and Gaussian mutational dragonfly algorithm (QGDA) is utilized as a searching core in a wrapper feature selection technique, and it is compared with other advanced feature selection methods. The results show that QGDA achieves substantial superiority in feature selection through optimum fitness and minimum error rate. Also, the results of QGDA on the three classical engineering design problems have demonstrated that the proposed method can effectively solve these constraints problems. It is encouraging that the proposed method can be used as a useful, auxiliary tool for solving complex optimization problems.

The sine cosine algorithm (SCA) was proposed for solving optimization tasks, of which the way to obtain the optimal solution is mainly through the continuous iteration of the sine and cosine update formulas. However, SCA also faces low population diversity and stagnation of locally optimal solutions. Hence, we try to eliminate these problems by proposing an enhanced version of SCA, named ESCA_PSO. ESCA_PSO is proposed based on hybrid SCA and particle swarm optimization (PSO) by incorporating multiple mutation strategies into the original SCA_PSO. To validate the effect of ESCA_PSO in handling global optimization problems, ESCA_PSO was compared with quality algorithms on various types of benchmark functions. In addition, the proposed ESCA_PSO was employed to tune the best parameters of support vector machines for dealing with medical diagnosis tasks. The results prove the efficiency of the proposed algorithms in solving optimization problems.

The butterfly optimization algorithm (BOA) is a swarm intelligence optimization algorithm proposed in 2019 that simulates the foraging behavior of butterflies. Similarly, the BOA itself has certain shortcomings, such as a slow convergence speed and low solution accuracy. To cope with these problems, two strategies are introduced to improve the performance of BOA. One is the random replacement strategy, which involves replacing the position of the current solution with that of the optimal solution and is used to increase the convergence speed. The other is the crisscross search strategy, which is utilized to trade off the capability of exploration and exploitation in BOA to remove local dilemmas whenever possible. In this case, we propose a novel optimizer named the random replacement crisscross butterfly optimization algorithm (RCCBOA). In order to evaluate the performance of RCCBOA, comparative experiments are conducted with another nine advanced algorithms on the IEEE CEC2014 function test set. Furthermore, RCCBOA is combined with support vector machine (SVM) and feature selection (FS)—namely, RCCBOA-SVM-FS—to attain a standardized construction model of overseas Chinese associations. It is found that the reasonableness of bylaws; the regularity of general meetings; and the right to elect, be elected, and vote are of importance to the planning and standardization of Chinese associations. Compared with other machine learning methods, the RCCBOA-SVM-FS model has an up to 95% accuracy when dealing with the normative prediction problem of overseas Chinese associations. Therefore, the constructed model is helpful for guiding the orderly and healthy development of overseas Chinese associations.

COVID-19 is currently raging worldwide, with more patients being diagnosed every day. It usually is diagnosed by examining pathological photographs of the patient's lungs. There is a lot of detailed and essential information on chest radiographs, but manual processing is not as efficient or accurate. As a result, how efficiently analyzing and processing chest radiography of COVID-19 patients is an important research direction to promote COVID-19 diagnosis. To improve the processing efficiency of COVID-19 chest films, a multilevel thresholding image segmentation (MTIS) method based on an enhanced multiverse optimizer (CCMVO) is proposed. CCMVO is improved from the original Multi-Verse Optimizer by introducing horizontal and vertical search mechanisms. It has a more assertive global search ability and can jump out of the local optimum in optimization. The CCMVO-based MTIS method can obtain higher quality segmentation results than HHO, SCA, and other forms and is less prone to stagnation during the segmentation process. To verify the performance of the proposed CCMVO algorithm, CCMVO is first compared with DE, MVO, and other algorithms by 30 benchmark functions; then, the proposed CCMVO is applied to image segmentation of COVID-19 chest radiography; finally, this paper verifies that the combination of MTIS and CCMVO is very successful with good segmentation results by using the Feature Similarity Index (FSIM), the Peak Signal to Noise Ratio (PSNR), and the Structural Similarity Index (SSIM). Therefore, this research can provide an effective segmentation method for a medical organization to process COVID-19 chest radiography and then help doctors diagnose coronavirus pneumonia (COVID-19).

Intradialytic hypotension (IDH) is a serious complication of hemodialysis (HD), with an incidence of more than 20%. IDH induces ischemic organ damage and even reduces the ultrafiltration and duration of HD sessions. Frequent attacks of IDH are a risk factor for death in HD patients. Malnutrition is common in HD patients and is also associated with mortality. Although the link between IDH episodes and malnutrition has been observed in practice, it has not been supported by the data. To study the relationship, we propose a promising hybrid model called BSCWJAYA_KELM, which is a wrapper feature selection method based on a variant of the JAYA optimization algorithm (SCWJAYA) and Kernel extreme learning machine (KELM). In this paper, we verify the optimization capability of the SCWJAYA algorithm in the model by comparing experiments with some state-of-the-art methods for IEEE CEC2014, IEEE CEC2017, and IEEE CEC2019 benchmark functions. The prediction accuracy of BSCWJAYA_KELM is validated by the public datasets and the HD dataset. In the experiments on the HD dataset, 1940 HD sessions of 178 HD patients are analyzed by the developed BSCWJAYA_KELM model. The key indicators selected from vast amounts of data are serum uric acid, dialysis vintage, age, diastolic pressure, and albumin. The BSCWJAYA_KELM method is a stable and excellent prediction model that can achieve a more accurate prediction of IDH.

This study proposed an improved sine–cosine algorithm (SCA) for global optimization tasks. The SCA is a meta-heuristic method ground on sine and cosine functions. It has found its application in many fields. However, SCA still has some shortcomings such as weak global search ability and low solution quality. In this study, the chaotic local search strategy and the opposition-based learning strategy are utilized to strengthen the exploration and exploitation capability of the basic SCA, and the improved algorithm is called chaotic oppositional SCA (COSCA). The COSCA was validated on a comprehensive set of 22 benchmark functions from classical 23 functions and CEC2014. Simulation experiments suggest that COSCA’s global optimization ability is significantly improved and superior to other algorithms. Moreover, COSCA is evaluated on three complex engineering problems with constraints. Experimental results show that COSCA can solve such problems more effectively than different algorithms.

As a typical nature-inspired swarm intelligence algorithm, because of the simple framework and good optimization performance, salp swarm algorithm (SSA) has been extensively applied to a lot of practical problems. Nevertheless, when facing a number of complicated optimization problems, particularly the high dimensionality and multi-dimensional problems, SSA will come to stagnation and decrease the optimal performance. To tackle this problem, this paper presents an enhanced SSA (ESSA) in which several strategies, including orthogonal learning, quadratic interpolation, and generalized oppositional learning are embedded to boost the global exploration and local exploitation performance of SSA. Orthogonal learning can help the worse salp break away from local optima, while quadratic interpolation is utilized to improve the accuracy of the global optimal through local search near the globally optimal solution. Also, generalized oppositional learning is used to improve the population quality through the initialization step and generation jumping. These strategies work together to assist SSA in promoting convergence performance. At the last CEC2017 benchmark suite and CEC2011, a real-world optimization benchmark is employed to estimate the property of ESSA in dealing with the high dimensionality and multi-dimensional problems. Three constrained engineering optimization problems are also used to assess the capability of ESSA in tackling practical engineering application problems. The experimental results and responding analysis make clear that the presented algorithm significantly outperforms the original SSA and other state-of-the-art methods.

This paper focuses on the study of Coronavirus Disease 2019 (COVID-19) X-ray image segmentation technology. We present a new multilevel image segmentation method based on the swarm intelligence algorithm (SIA) to enhance the image segmentation of COVID-19 X-rays. This paper first introduces an improved ant colony optimization algorithm, and later details the directional crossover (DX) and directional mutation (DM) strategy, XMACO. The DX strategy improves the quality of the population search, which enhances the convergence speed of the algorithm. The DM strategy increases the diversity of the population to jump out of the local optima (LO). Furthermore, we design the image segmentation model (MIS-XMACO) by incorporating two-dimensional (2D) histograms, 2D Kapur's entropy, and a nonlocal mean strategy, and we apply this model to COVID-19 X-ray image segmentation. Benchmark function experiments based on the IEEE CEC2014 and IEEE CEC2017 function sets demonstrate that XMACO has a faster convergence speed and higher convergence accuracy than competing models, and it can avoid falling into LO. Other SIAs and image segmentation models were used to ensure the validity of the experiments. The proposed MIS-XMACO model shows more stable and superior segmentation results than other models at different threshold levels by analyzing the experimental results.

The effective analytical processing of pathological images is crucial in promoting the development of medical diagnostics. Based on this matter, in this research, a multi-level thresholding segmentation (MLTS) method based on modified different evolution (MDE) is proposed. The MDE is the primary benefit offered by the suggested MLTS technique, which is a novel proposed evolutionary algorithm in this article with significant convergence accuracy and the capability to leap out of the local optimum (LO). This optimizer came into being mostly as a result of the incorporation of the movement mechanisms of white holes, black holes, and wormholes into various evolutions. Thus, the developed MLTS approach may provide high-quality segmentation results and is less susceptible to segmentation process stagnation. To validate the efficacy of the presented approaches, first, the performance of MDE is validated using 30 benchmark functions, and then the proposed segmentation method is empirically compared with other comparable methods using standard pictures. On the basis of breast cancer and skin cancer pathology images, the developed segmentation method is compared to other competing methods and experimentally validated in further detail. By analyzing experimental data, the key compensations of MDE are proven, and it is experimentally shown that the unique MDE-based MLTS approach can achieve good performance in terms of many performance assessment indices. Consequently, the proposed method may offer an efficient segmentation procedure for pathological medical images.

Paraquat (PQ) poisoning seriously harms the health of humanity. An effective diagnostic method for paraquat poisoned patients is a crucial concern. Nevertheless, it's difficult to identify the patients with low intake of PQ or delayed treatment. Here, a new efficient diagnostic approach to integrate machine learning and gas chromatography-mass spectrometry (GC-MS), named GEE, is proposed to identify the PQ poisoned patients. First, GC-MS provides the original data that efficiently identified the paraquat-poisoned patients. According to the high dimensionality of the original data, in the second stage, the chaos enhanced grey wolf optimization (EGWO) is adopted to search the optimal feature sets to improve the accuracy of identification. Finally, the extreme learning machine (ELM) is used to identify the PQ poisoned patients. To efficiently evaluate the proposed method, four measures were used in our experiments and comparisons were made with six other methods. The PQ-poisoned patients and robust volunteers can be well identified by GEE and the values of AUC, accuracy, sensitivity and specificity were 95.14%, 93.89%, 94.44% and 95.83%, respectively. Our experimental results demonstrated that GEE had better performance and might serve as a novel candidate diagnosis of PQ-poisoned patients.

Intradialytic hypotension (IDH) is the most common acute complication in hemodialysis (HD) sessions and is associated with increased morbidity and mortality in HD patients. To prevent the episode of IDH, it is critical to predict its occurrence. Chronic kidney disease-mineral and bone disorders (CKD-MBD) induce cardiac and vascular calcification, which impairs the compensatory mechanisms of blood pressure during HD. In this study, we proposed a feature selection framework called BSWEGWO_KELM to analyze 1940 records from 178 HD patients, which was based on an enhanced grey wolf optimization (GWO) algorithm and the kernel extreme learning machine (KELM). Then, global optimization experiments, together with feature selection experiments on public data sets and HD dataset, were performed to verify the effectiveness of the BSWEGWO_KELM method. The experimental results showed that the established BSWEGWO_KELM had the capability of screening out the key indicators such as dialysis vintage, mean arterial pressure (MAP), alkaline phosphatase (ALP), and intact parathyroid hormone (iPTH). Consequently, BSWEGWO_KELM can be applied as a practical and accurate method to predict IDH.

The slime mould algorithm (SMA) is a logical swarm-based stochastic optimizer that is easy to understand and has a strong optimization capability. However, the SMA is not suitable for solving multimodal and hybrid functions. Therefore, in the present study, to enhance the SMA and maintain population diversity, a dispersed foraging SMA (DFSMA) with a dispersed foraging strategy is proposed. We conducted extensive experiments based on several functions in IEEE CEC2017. The DFSMA were compared with 11 other meta-heuristic algorithms (MAs), 10 advanced algorithms, and 3 recently proposed algorithms. Moreover, to conduct more systematic data analyses, the experimental results were further evaluated using the Wilcoxon signed-rank test. The DFSMA was shown to outperform other optimizers in terms of convergence speed and accuracy. In addition, the binary DFSMA (BDFSMA) was obtained using the transform function. The performance of the BDFSMA was evaluated on 12 data sets in the UCI repository. The experimental results reveal that the BDFSMA performs better than the original SMA, and that, compared with other optimization algorithms, it improves classification accuracy and reduces the number of selected features, demonstrating its practical engineering value in spatial search and feature selection.

In this paper, we report a sensitive refractive index sensor based on the titanium dioxide (TiO $ _2$ ) film coated excessively tilted fiber grating (Ex-TFG) operating around the dispersion tuning point. The effects of the TiO $ _2$ film on the cladding mode field, transmission, and sensing characteristics of the Ex-TFG are outlined. It is found that the TiO $ _2$ film induces a two-step mode transition process for both TE/TM $ _{0,l}$ and HE/EH $ _{v,l}$ ( $ v=1,2,\ldots$ ) cladding modes. During the first transition, the same optimized film thickness can be obtained for TE $ _{0,l}$ , HE $ _{1,l}$ , and EH $ _{2,l-1}$ cladding modes. The results show that the sensitivity can be optimized based on the combined effects of the intermodal separation, initial wavelength separation, and the variation in the phase matching curve (i.e., $ \partial \Lambda / \partial \lambda$ ). The improved sensitivities are up to $ 32261.2\;\text{nm}/\text{RIU}$ and $ 23183.3\;\text{nm}/\text{RIU}$ by tracking dual resonance band corresponding to the s-polarized TE $ _{0,10}$ and p-polarized EH $ _{2,9}$ cladding modes, respectively, in the low refractive index region. For the single resonance corresponding to the s-polarized HE $ _{1,9}$ cladding mode, the sensitivity as high as $ 25860.0\;\text{nm}/\text{RIU}$ is achieved. The sensitivity is significantly enhanced by a factor of $ 13.2 \sim 18.1$ and $ 3.7 \sim 5.2$ times as compared with the uncoated Ex-TFG and classical TiO $ _2$ film coated long period fiber grating with reduced fiber diameter, respectively. The sensitivity can be further improved, which makes the film coated Ex-TFG very attractive for the refractive index sensing.

In this work, theoretical design of a novel band pass filter based on the turning-around-point long period fiber grating (TAP-LPFG) having cladding index modulation is proposed and demonstrated. Cladding modes are excited by introducing the incident light into the fiber cladding of the input end of the TAP-LPFG. At the grating region, only those cladding modes that satisfy the phase matching condition couple back to the fiber core to obtain the band pass transmission. It is shown that a band pass filter with –3 dB width of from ~115 to ~198 nm with the center wavelength of ~1554.4 nm can be obtained, and the suppression of more than 10 dB for the side-lobes is achieved in a wide wavelength range. This novel TAP-LPFG based device can find applications in optical communications for wide band pass filter.