An Analysis of Machine Learning Algorithms for Early Prediction of Heart Attack in Stroke Patients
Abstract
Predicting heart attacks early in stroke patients through data analysis is crucial to reducing the high mortality rate associated with these conditions. However, accurately predicting heart attacks in stroke patient data poses a challenge. Early detection of stroke-related diseases is beneficial for prevention or early intervention. Machine learning and data mining play important roles in stroke prediction. This paper proposes an effective method for identifying stroke and compares the performance of three machine learning algorithms: Decision Tree, Naïve Bayes and K-Nearest Neighbors. The study analyzes the performance of these algorithms on a stroke dataset. The preliminary results demonstrate that the Decision Tree algorithm achieves the highest accuracy of 95.75%, outperforming the Naïve Bayes and K-Nearest Neighbors algorithms.
Keywords
Download Options
Introduction
A stroke refers to the interruption of blood supply to the brain caused by the rupture of one or more small blood vessels. This interruption deprives the brain of essential oxygen and nutrients, leading to neuronal death. Ischemic stroke, the most common type of stroke, occurs when a blood vessel supplying the brain becomes blocked by a blood clot, often due to factors such as smoking and high cholesterol [1][5]. While many people believe that strokes primarily affect older individuals due to their higher risk of heart disease and diabetes, strokes are caused by blockages in the brain's blood vessels that carry oxygen and nutrients. This arterial dysfunction can be caused by various factors, including smoking, diabetes, high blood pressure, cholesterol, and heart disease, with variations based on age and gender.
The clinical symptoms of stroke include sudden collapse, mental coma, impaired speech, and hemiplegia [6]. A heart attack is myocardial decay caused by acute and persistent ischemia and hypoxia of the coronary artery, presenting symptoms such as arrhythmia, shock, or heart failure, which can be fatal [9]. Stroke complicated by a heart attack refers to a cerebral infarction accompanied by a heart attack. It is reported that 30% of stroke cases are complicated by a heart attack, with a high mortality rate of 54% [5]. The main causes of death are ventricular arrhythmia, acute left heart failure, and cardiogenic shock. Troponin is an effective marker for detecting heart attacks [9], commonly used in clinics. However, a drawback of troponin is that it starts changing only four hours after a heart attack has occurred, resulting in a time delay for detection. In contrast, heart attacks have a rapid onset, and sudden deaths can occur easily in heart attack patients.
Conclusion
In conclusion, the results of this study demonstrate the effectiveness of machine learning algorithms in predicting heart stroke. These algorithms can serve as valuable tools for healthcare professionals in making informed decisions and implementing preventive strategies. Continued advancements in machine learning techniques can contribute to further improvements in heart stroke prediction and ultimately lead to better patient outcomes.
Based on the experimental results, it can be observed that all three machine learning algorithms showed promising performance in predicting heart stroke. The Decision Tree algorithm achieved the highest accuracy, precision, and recall scores, followed closely by Naïve Bayes. K-Nearest Neighbors (KNN) algorithm exhibited slightly lower performance compared to the other two algorithms.