Cerebrovascular Stroke Detection Using Machine Learning Algorithms

• Published in: 2022 1st International Conference on Computational Science and Technology (ICCST) pp. 273 - 275
• Main Authors: Susitra, K., T, Mahalakshmi, E, Madhumitha
• Format: Conference Proceeding
• Language: English
• Published: IEEE 09.11.2022
• Subjects: Aging; Cerebrovascular; Classification algorithms; Forestry; Kaggle; Machine learning algorithms; Naive Bayes methods; Naϊve Bayes Algorithm; Prediction algorithms; Random Forest Algorithm; Scientific computing; Stroke
• DOI: 10.1109/ICCST55948.2022.10040416

In terms of mortality rates, strokes are considered to be the second most common reason for death. Every year, 15 million individuals worldwide have a stroke, and of those,5 million perish and another 5 million become permanently crippled. The prevention and detection of stroke have become increasingly important in order to avoid its devastating effects. Using the Kaggle stroke dataset, health records are analyzed. A list of attributes is obtained from the health records: aging, BMI, gender, glucose level, smoking, blood sugar, etc ... The information gathered is used as inputfor Navies Bayesian and Random Forest classification to predict stroke. Naive Bayes classifiers categorize data using the principles of probability theory. Bayes' theorem is the foundation of the naive Bayes algorithm. A random forest algorithm builds decision trees from data samples and utilizes each tree's forecast to decide which is the best course of action. Using these algorithms, the best method to predict cerebrovascular strokes is determined by the accuracy and precision of the prediction.