Publication

Abstract : The diagnosis of cardiac disease is a time-consuming procedure. The presence and magnitude of cardiovascular disease are revealed by the morphology of the electrocardiogram (ECG) sensor signal. The scarcity of instances of the irregular condition collected using ECG sensors is the most significant factor limiting the identification of cardiac disease. In this study, we use a Chebyshev function, also known as Chebfun, to solve the problem of cardiac disease detection with fewer ECG sensor signals. The Chebfun uses its coefficients to approximate the signal. Instead of the traditional handcrafted features, these coefficients, known as Chebfun coefficients, are used as features. With these features of Chebfun, the present work aims to give an interpretation of the features learned by the model for cardiac disease detection. The proposed work makes use of a machine learning algorithm to tackle the problem of better performances using fewer ECG sensor signals. Machine learning algorithms such as SVM, logistic regression, decision tree, and AdaBoost are used in this study.