Scholarlink

<p>The aim of current work is to perform the numerical solutions of the fractional order nonlinear mathematical lungs cancer model by applying a stochastic computing approaches. The fractional kind of the derivatives is used to present the reliable and accurate performance of the lungs cancer model in comparison with the integer order derivatives. The lungs cancer model is categorized into the cells of cancer in lung tissue, cancer cells spread to the remaining body parts, and cells of immune in lung tissue. Three cases of the fractional order derivatives between 0 and 1 have been used to present the numerical solutions of the model in order to check, which fractional values perform better closer to 0 or 1. A construction of single hidden layer is performed by applying the radial basis activation function, whereas the tests of optimization are provided through the Bayesian regularization. A dataset is collected through the Runge–Kutta solver, which is used to reduce the mean square error by data separating into training, verification and testing with reasonable percentages. The proposed stochastic computing process contains a radial basis merit function, twenty neurons, and the optimization through the Bayesian regularization. The correctness of proposed solver is perceived via outputs overlapping, and minor absolute error. Moreover, the reliability of scheme is approved through different tests including state transition, regression coefficient performances, and error histogram values.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Solution of fractional order mathematical lungs cancer operation model: a radial basis neural network

  • 期刊论文

  • 发表日期：

    2025年6月17日

查看全文
  • Zulqurnain Sabir,
  • M. A. Abdelkawy,
  • Muhammad Asif Zahoor Raja,
  • Warda Mushtaq

摘要

The aim of current work is to perform the numerical solutions of the fractional order nonlinear mathematical lungs cancer model by applying a stochastic computing approaches. The fractional kind of the derivatives is used to present the reliable and accurate performance of the lungs cancer model in comparison with the integer order derivatives. The lungs cancer model is categorized into the cells of cancer in lung tissue, cancer cells spread to the remaining body parts, and cells of immune in lung tissue. Three cases of the fractional order derivatives between 0 and 1 have been used to present the numerical solutions of the model in order to check, which fractional values perform better closer to 0 or 1. A construction of single hidden layer is performed by applying the radial basis activation function, whereas the tests of optimization are provided through the Bayesian regularization. A dataset is collected through the Runge–Kutta solver, which is used to reduce the mean square error by data separating into training, verification and testing with reasonable percentages. The proposed stochastic computing process contains a radial basis merit function, twenty neurons, and the optimization through the Bayesian regularization. The correctness of proposed solver is perceived via outputs overlapping, and minor absolute error. Moreover, the reliability of scheme is approved through different tests including state transition, regression coefficient performances, and error histogram values.