Stroke (Cerebrovascular Accident - CVA) causes significant damage due to the interruption or reduction of blood supply to an area of the brain. This condition can result in severe sequelae, including cognitive impairment, paralysis, speech and coordination difficulties, directly affecting the patient's quality of life. The brain damage resulting from a stroke can have irreversible impacts on physical and mental capacity, highlighting the importance of preventive measures, rapid interventions, and rehabilitation to minimize adverse consequences. In this study, we propose a method for monitoring and predicting strokes that integrates Internet of Things (IoT) devices for remote patient monitoring. Recognizing the severity of stroke-associated sequelae, our approach aims to mitigate adverse impacts through preventive measures and timely interventions. Using a combination of machine learning algorithms, including Naive Bayes, Multilayer Perceptron, Support Vector Machine, k-Neighbors, Decision Tree, XGBoost, and Random Forest, we aim to assess the risk of stroke occurrence, with XGBoost standing out with an Accuracy of 98.52% and a testing time of 0.076ms.