A NOVEL PROBABILISTIC ARTIFICIAL NEURAL NETWORKS APPROACH FOR DIAGNOSING HEART DISEASE
Authors/Creators
- 1. Department of Computer Engineering, Shabestar Branch, Islamic Azad University, Shabestar, Iran
- 2. Executive Master of Business Administration (EMBA) Student, Sanandaj Branch, Islamic Azad University, Sanandaj, Iran
- 3. Department of Computer Engineering, Tabriz Branch, Islamic Azad University, Tabriz, Iran
- 4. Department of Computer Engineering, Shabestar Branch, Islamic Azad University, Shabestar, Iran 2 Executive Master of Business Administration (EM
- 5. 1Department of Computer Engineering, Shabestar Branch, Islamic Azad University, Shabestar, Iran
Description
In today's world one of the most common diseases are heart disease which its mortality and disability is high. Therefore, heart disease is one of the biggest health problems in the world. Since the diagnosis of heart disease in people is very important, a method should be used in the right diagnosis of heart diseases that have the least errors in heart disease diagnosis. For this reason, in this paper, Probabilistic Neural Networks (PNNs) for the diagnosis of heart disease from a dataset that includes 303 samples from different patients is used. In this paper, we have implemented PNN in the MATLAB environment. As well as, the efficiency criteria in this paper is to maximize accuracy of heart disease diagnosis in the process of training and testing. According to the Cleveland dataset which contains 303 samples, we found that the accuracy of training and test accuracy are 87% and 75% respectively.