Efficient Prediction of Brain Stroke Using Machine Learning

Abstract

In recent years strokes are one of the leading causes of death by affecting the central nervous system. Among different types of strokes, ischemic and hemorrhagic majorly damages the central nervous system. According to the World Health Organization (WHO), globally 3% of the population are affected by subarachnoid hemorrhage, 10% with intracerebral hemorrhage, and the majority of 87% with ischemic stroke. In this research work, Machine Learning techniques are applied in identifying, classifying, and predicting the stroke from medical information. The existing research is limited in predicting risk factors pertained to various types of strokes. To address this limitation a Stroke Prediction (SPN) algorithm is proposed by using the improvised RNN in analyzing the levels of risks obtained within the strokes. This research of the Stroke Predictor (SPR) model using machine learning techniques improved the prediction accuracy is higher when compared with the existing models. In our work we will be using the following algorithms such as Convolution Neural Network (CNN) as existing and Recurrent Neural Network (RNN) as proposed and its accuracy is been calculated and well compared. From the results obtained it is proved that proposed Recurrent Neural Network (RNN) works better than existing Convolution Neural Network (CNN)..