Support Vector Machine Based Classification for Telemedicine Based on ECG Data

Heart disease is a significant global health concern, and early detection plays a crucial role in improving patient outcomes. This study proposes an electrocardiography (ECG) based heart disease detection using a support vector machine (SVM) machine learning model. The SVM model is trained on a comprehensive dataset consisting of ECG recordings from individuals with and without heart disease allowing it to be effectively trained to differentiate between the two. The process begins with preprocessing the ECG data to eliminate noise and artifacts followed by feature extraction to capture relevant characteristics of the heart signal. These features include time-domain and frequency-domain measures such as QRS complex duration R peak amplitude and spectral components. The SVM model is then trained using an appropriate kernel function to map the feature space enabling efficient classification. The performance of the model is evaluated using accuracy, true positive, false positive, sensitivity and specificity. The proposed SVM model achieves high accuracy in detecting heart disease demonstrating the effectiveness of ECG-based analysis and machine learning. The system shows promise in assisting healthcare professionals in identifying heart disease at an early stage which can potentially lead to timely intervention and improved patient outcomes.