Noise Reduction of Biomedical Signal using Artificial Neural Network Model

The Electromyography (EMG) is a very important biomedical signal and can be used for verity of applications in clinical or biomedical field. This signal is used to detect abnormal muscle activities like impaired nourishment of an organ or part of body, inflammation of muscles, pinched nerves and peripheral nerve damages etc. The EMG signal is controlled by nervous system and is dependent on the anatomical and physiological properties of muscles and is affected due to artifacts. Therefore the EMG signal is complicated signal and noise-prone. This noise signal reduces the performance of EMG signal. During signal processing, the system picks up noise signal along with desired signal. In this paper, Artificial Intelligent model using Focused Time Lagged Recurrent Neural Network with a single hidden layer has been developed. From the implication of findings, FTLRNN reduces noise intelligently from the EMG signal. The difference between EMG with noise and desired EMG signal is computed from the performance measures MSE, NMSE and r.