A Qualitative Analysis of Independent Component Analysis Based Algorithms for the Removal of Artifacts from Electroencephalography Signals

Abstract

Problem statement: Independent Component Analysis (ICA) based algorithms applied in the context to remove the artifacts from the EEG signals are evaluated with appropriate metrics and it compares and contrasts the performance of the different methods for such applications. The primary goal is to gain some insight into relative performance of the various methods. Approach: CA is a statistical and computational technique for revealing hidden factors that underlie sets of random signals. In the ICA model the data samples are assumed to be linear mixture of some unknown latent variables and the mixing system is also unknown. The latent variables are assumed to have a nongaussian distribution. These variables are the independent components of the observed data which can be found, up to some degree of accuracy, using different algorithms based on ICA techniques. Results: The algorithms based on ICA with different approaches to be considered are JADE, Fast ICA, infomax and extended infomax and these performances are measured in terms of Entropy, PSNR and Speed. The simulation results show that the performance of each algorithm is to be preferred over another in terms of speed and reliability. A framework for accommodating four ICA algorithms is developed and hence selects the best algorithm for the specific type of data. Conclusion: ICA plays a vital role in removing of artifacts in EEG signals .It maintains the similarity in their patterns when subject is performing the mental task. The traditional methods applied for remove artifacts can only compromise between eliminating artifacts and protecting useful signals so that the result is not very satisfying. ICA method can protect the useful signals as well as obviously weaken even entirely remove the artifacts in multichannel EEG signals, this characteristic of ICA is the key to get stable EEG patterns which can be used for mental task classification.