A hybrid swarming computing approach to solve the biological nonlinear Leptospirosis system. (August 2022)

• Record Type: Journal Article
• Authors: Botmart, Thongchai; Sabir, Zulqurnain; Asif Zahoor Raja, Muhammad; weera, Wajaree; Ali, Mohamed R.; Sadat, R.; Aly, Ayman A.; Alosaimy, ; Saad, Ali
• Is Part Of: Biomedical signal processing and control. Volume 77(2022)
• Publication Date: 2022-08
• Subjects: Leptospirosis -- Artificial neural networks -- Neuron analysis -- Local/global search methods -- Statistical analysis -- Reference solutions; Signal processing -- Periodicals; Biomedical engineering -- Periodicals; Signal Processing, Computer-Assisted -- Periodicals; Image Processing, Computer-Assisted -- Periodicals; Biomedical Engineering -- Periodicals; 610.28
• DOI: 10.1016/j.bspc.2022.103789
• Languages: English
• ISSNs: 1746-8094

Abstract

Highlights: The numerical simulations of the LDM are presented based on the ANNs along with the optimization procedures of PSO-SQP. The neuron analysis is presented in terms of absolute error (AE) by taking small and large number of neurons for the solutions of the LDM. The exactness of the proposed stochastic procedures based swarming intelligent paradigms are observed using the comparison of the obtained and reference results. The accuracy of the stochastic computing schemes based swarming procedures is adjudicated to find the small values of the AE for the LDM. The stability of the stochastic procedures based swarming intelligent paradigms are observed using the statistical mean square error (MSE), semi-interquartile (SIR), variance account for (VAF) and Theil's inequality coefficient (TIC). Abstract: This study indicates the design of swarming procedure based on the stochastic framework of artificial neural networks (ANNs) along with the particle swarm optimization (PSO) and sequential quadratic programming (SQP) for the Leptospirosis disease model (LDM). LDM is zoonotic disease, which broadly occurs in each continent of the world. LDM is dependent upon three classes and the numerical solutions are presented by using the procedures of ANNs-PSO-SQP. The construction of a merit function is provided based on the LDM and then optimized by using the PSO-SQP. The proposed ANNs-PSO-SQP scheme is used to LDM to indorse the exactness, precision, trustworthiness, and aptitude of the ANNs-PSO-SQP. The obtained ANNs-PSO-SQP of the LDM compared with the Adams method, which confirm the significance of the proposed ANNs-PSO-SQP. The neuron analysis based on the larger and smaller neurons will be provided to authenticate the correctness of the ANNs-PSO-SQP for solving the LDM. Moreover, statistical representations based on different operators will be provided to check the reliability of the ANNs-PSO-SQP for solving the LDM.