Efficient classification and segmentation of brain stroke mri images using enhanced convolutional neural networks

Abstract

newline In today s world, stroke is a life-threatening disease requiring newlineimmediate medical attention to prevent permanent brain damage or death. newlineWorldwide, strokes rank second among the top causes of death. When there is newlinea difficulty with blood flow to a certain area of the brain, it happens when a newlineblood vessel is blocked or when the brain is bleeding. Anybody can have a newlinestroke, from children to adults and, more common in people over 65. According newlineto WHO research, 15 million people worldwide suffer from strokes yearly, of newlinewhich 5 million deaths occur and more than 5 million become permanently newlinedisabled. In persons under 40, high blood pressure is the main contributor to newlinestroke. However, around 8% of children also experience strokes. The greatest newlinecause of disability, particularly in low- and middle-income nations, is stroke. newlineThis research focuses on enhancing stroke diagnosis through newlineadvanced medical image processing techniques, addressing the limitations of newlineexisting methods. The diagnosis of brain diseases has been made by medical newlineimaging methods such as CT, MRI, and PET. Compared to CT, Magnetic newlineresonance imaging (MRI) is a valuable tool to identify many vascular and brain newlineabnormalities, and it is capable of producing multiplanar images that help to newlinevisualize the brain in different orientations, providing a more comprehensive newlineunderstanding of the location and extent of brain damage caused by ischemic newlineand hemorrhagic stroke.