EARLY BRAIN STROKE PREDICTION USING MACHINE LEARNING ALGORITHMS

MACHINE LEARNING

A brain stroke is a serious medical emergency that can have life-altering or permanently crippling effects. Cell death due to inadequate blood supply to the brain causes it to happen. In India, the annual incidence of stroke stands at 141 per 100,000 people. Predicting early brain strokes has become increasingly challenging, requiring time-consuming assessments. Given the life-or-death nature of stroke diagnoses and prognoses, precision and accuracy are crucial. Machine learning techniques offer a means to predict stroke issues by analyzing extensive medical data. By leveraging a substantial dataset for training and testing, the study assesses the predictive capabilities of various machine learning methods. A few examples of these techniques are K-nearest neighbour, decision trees, logistic regression, SVMs, Random Forest, AdaBoost, and Bernoulli navie bayes. The research assesses the efficacy of the model by means of the F1 score, Accuracy, Precision, and Recall, which are extracted from the confusion matrix. A web application will be developed, enabling users to input relevant parameters. Using this Flask-based application, the model processes these parameters. This approach, powered by the most accurate and effective method, can predict the likelihood of strokes.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.