BRAIN STROKE DETECTION AND MULTI-DISEASE PREDICTION SYSTEM

In recent years, early and accurate diagnosis of brain stroke and multiple chronic diseases has become a critical challenge in the healthcare field. This research presents an integrated deep learning and machine learning framework designed for brain stroke detection and multi-disease prediction. The proposed system consists of two major components: an image processing part and a textual data analysis part. In the image module, medical images are preprocessed by resizing and converting them from RGB to grayscale to enhance feature clarity. The dataset is then divided into training and testing sets and trained using a Convolutional Neural Network (CNN) for effective brain stroke detection. Meanwhile, the textual module processes patient health records by cleaning and removing noisy data. The refined dataset is split and trained using a Support Vector Machine (SVM) algorithm to classify and predict multiple diseases. By integrating CNN and SVM techniques, the system leverages the strengths of both deep learning and traditional machine learning methods—CNN for extracting visual features and SVM for analyzing textual data. This hybrid approach enhances prediction accuracy, reduces processing time, and provides a comprehensive diagnostic tool to assist healthcare professionals in early disease identification nand clinical decision-making.