Smart Stroke Detection: Cutting-Edge Machine Learning and Optimized Algorithms for Early Diagnosis
Authors:-Mr. N.V.S Gopalam, K.Tanoosh, Ch.Sowjanya, Y.Navatej, K.Banny, B.Lakshmi Jahnavi
Abstract-Stroke, which ranks as the second leading cause of death worldwide, requires prompt and precise prediction for effective intervention. This research investigates the use of advanced machine learning techniques to improve stroke prediction models. Initially, classifiers such as Decision Tree (DT), Random Forest (RF), and Support Vector Machine (SVM) were applied, followed by the incorporation of advanced algorithms like Extreme Gradient Boosting (XGBoost) and Light Gradient Boosting Machine (LGBM) to enhance predictive accuracy. Various evaluation metrics, including accuracy, sensitivity, error rates, and log loss, were employed to assess the performance of the models. The findings demonstrate the effectiveness of machine learning algorithms, with XGBoost achieving an impressive accuracy rate of 98%. Additionally, LGBM played a significant role in boosting overall accuracy. These results highlight the critical contribution of advanced machine learning techniques to enhancing stroke prediction. By leveraging these state-of-the-art predictive models, the study advocates for their integration into clinical settings, aiming to expedite accurate diagnoses, improve patient care, and advance stroke detection capabilities. Keywords: Brain Stroke, Machine Learning, Light Gradient Boosting Machine (LGBM), Extreme Gradient Boosting (XGBoost).
