Evaluate and determine the most effective treatment parameters in esophageal cancer using intelligent systems

Abstract

In recent years, use of the artificial neural networks has been considered in predicting the effects of different variables on a given variable and modeling these variables have with one another. In this research, first, artificial neural networks have been used to predict the results of treatment of esophageal cancer in patients with esophageal squamous cell carcinoma using chemotherapy, radiotherapy and then Nyvajvnt surgery. In addition, the Particle Swarm Optimization (PSO) is used for training the neural network. Then, using the combined neural network and genetic algorithms, a method is proposed to select the most effective treatment parameters among a set of factors affecting the proposed treatment process. Implementation results show that neural network can predict the level of satisfactory treatment of the cancer process. The results of methods for selecting the most effective parameters on the process of treatment among sixteen proposed parameters are compatible with the previous findings.

4 Figures and Tables

Showing 1-8 of 8 references

Comparison of artificial neural network models and Cox regression to predict survival in patients with gastric cancer

  • A Kazemnejad, E Hajizadeh, A Byglryan
  • 2010

A diagonally recurrent neural network approach to sensorless operation of the permanent magnet synchronous motor

  • T D Batzel, K Y Lee
  • 2000

Applying neural network models to prediction and data analysis in meteorology and oceanography

  • W W Hsieh, B Tang
  • 1998