Comparison of Linear Branching Programs and Neural Networks for ECG Classification

The electrocardiogram (ECG or EKG) is a diagnostic tool that measures and records the electrical activity of the heart in exquisite detail.  Interpretation of these details allows diagnosis of a wide range of heart conditions. Two methods used for the classification of electrocardiogram signals: the former based on linear branching program and the latter relying on neural networks. The linear branching programs approach  used to classify the Biomedical Signal  by means decision tree a decision tree in which the decision path is decided according to the value assumed by the projection of the input features. Neural Networks are well-know machine learning structures used in many different fields ranging for classification. Neural Networks have several degrees of freedom including number of hidden layers, neurons per hidden layer, and form of activation functions. In most cases, a two-layer Neural Network is sufficient to obtain a good classification. These two methods deals with all the requirements and difficulties related to working with electrocardiogram data. The results based on Normal ECG signal and Abnormalities of ECG signal. The proposed systems prove that carrying out complex tasks like ECG classification.