Performance Evaluation of Different Machine Learning Methods and Deep-Learning Based Convolutional Neural Network for Health Decision Making

Abstract

Now-a-days modern technology is used for health management and diagnostic strategy in the health sector. Machine learning usually helps in decision making for health issues using different models. Classification and prediction of disease are easily known with the help of machine learning techniques. The machine learning technique can be applied in various applications such as image segmentation, fraud detection, pattern recognition and disease prediction, etc. In the today’s world, maximum people are suffering from diabetes. The glucose factor in the blood is the main component of diabetes. Fluctuation of blood glucose level leads to diabetes. To predict the diabetes disease, machine learning and deep learning play major role which uses probability, statistics and neural network concepts, etc. Deep learning is the part of machine learning which uses different layers of neural network that decide classification and prediction of disease. In this chapter, we study and compare among different machine learning algorithms and deep neural networks for diabetes disease prediction, by measuring performance. The experiment results prove that convolution neural network based deep learning method provides the highest accuracy than other machine learning algorithms.