Novel ECG diagnosis model based on multi-stage artificial neural networks

Different ECG diagnosis tools are currently available, which include Artificial Neural Network-based ECG classifier application system. This paper proposes a novel model of multi-stage feed forward neural networks for ECG signal classification. The research is aimed at the design of an intelligent ECG diagnosis tool, which can recognise heart abnormalities while reducing the complexity, cost, and response time of the system. A number of neural network architectures are designed and compared for their abilities to classify six different heart conditions. The input data of the networks comprise 12 ECG features and 13 compressed components of each heart beat signal, which are obtained from the MIT/BIH database. Among the different architectures tested, a multi-stage network gave the highest recognition rate of 90.57%. This network is proposed as a suitable candidate to be used in intelligent ECG signal diagnosis systems.