Enhancement of QRS complex using a neural network based ALE

Describes the application of a neural network based adaptive line enhancer (ALE) for enhancement of the QRS complex corrupted with background noise. Fully connected modified recurrent neural network, which is the combined structure of Elman´s and Jordan´s RNNs with one-to-many connections, is used as a nonlinear adaptive filter in the ALE. The connecting weights between network nodes as well as the parameters of the node activation function such as gain, slope, and delay are updated at each iteration using the error backpropagation algorithm. The proposed network is firstly evaluated by performing linear and nonlinear system identification. The real ECG signal buried with moderate and severe background noise is applied to the ALE using a nonlinear neural network adaptive filter in order to enhance the weak QRS complex. It is verified that the proposed network is suitable for use in system identification. Simulation results also show that the neural network based ALE performs well the enhancement of the QRS complex from noisy ECG signals