Blood stroke Classification using Proposed CNN Model

Abstract

The faster medical treatment is provided, the better chances of recovery from a blood stroke. Early detection allows for prompt medical intervention, which can aid in the dissolution of the clot and the restoration of blood flow to the brain. This can minimize the damage caused by the stroke and reduce the risk of long-term disability or death. This study presents a proposed Convolutional Neural Network (CNN) model for the classification of blood stroke into two classes, blood clots or normal. For training the model, the Adam optimizer was used with a batch size of 32 and 220 epochs. The proposed model was evaluated using various performance metrics such as precision, recall, F1 score, and accuracy. The model had an overall accuracy of 92%, indicating that it can correctly classify cases of blood stroke. The findings of this study offer promising clues for the development of automated blood stroke detection systems based on deep learning models, which can help healthcare professionals make timely and accurate diagnoses.