Application of the rough set theory and BP neural network model in disease diagnosis

Application of the rough set theory and BP neural network model in disease diagnosis is discussed in this paper. BP neural network model was established, and trained by the real diagnosis data of nephritis, utilizing the neural network toolbox in Matlab software. In this way we were able to provide a good solution to the problem of diagnose for new patients based on their chemical test data. By data mining based on the rough set theory model, we further identified the key factors affecting the diagnosis, and obtained relatively high classification accuracy through validation under the BP neural network model.