A Comparative analysis to detect stroke using deep neural network, Recurrent neural network and KNN

Abstract

Stroke is the second leading cause of death, and stroke may be a long-lasting crippling illness. Stroke is a total mental failure exacerbated by a disruption in blood supply to the brain, or by blocking of blood flow to the head. According to the World Health Organisation, the mortality risk will begin to rise throughout the next year's stroke. Much research has been conducted on stroke disease identification. An artificial intelligence solution to stroke and its forms by in-depth education. Forms include Ischemical stroke, Hemorrhagic stroke, Acute Ischemic Attack. Databases from the research organization include obtained through our study. The system of preprocessing expels copied data, details lacking and inconsistent records. The key feature research technique is the estimation is used to decrease predictions and the application of deep learning measures whether or not the individual has chronic illness. In order to anticipate the stroke, classification by deep learning is revised. Once data are submitted, it tests on a qualified model and the predictions of multiple forms of stroke. This study focused primarily on a reliable means of forecasting stroke and specific styles of stroke.