Auditory evoked potential (AEP) recordings have been analyzed through independent component analysis (ICA) in the literature; however, the performance varies depending on the ICA algorithms used. There are very few studies that concentrate on the optimum parameter selection for estimating the AEP components reliably, while also recovering the specific artifact generated with the normal functioning of a cochlear implant (CI). The objective of this research is to determine which ICA algorithm, high-order statistics (HOS)-based or second-order statistic (SOS)-based, is more plausible to remove this artifact and estimate the AEP. The optimal parameters of three such ICA algorithms for estimating the components from a database of recordings were determined, and then the estimates for the AEP and CI artifact were compared using each method. All the algorithms estimate the CI artifact reasonably well, although only one SOS algorithm is better positioned to estimate the AEP; this is primarily because it uses the temporal structure of this signal as part of the ICA process.

Multi-channel Auditory Evoked Potentials (AEPs) are a useful methodology for evaluating the auditory performance of children with Cochlear Implants (CIs). These recordings are generally contaminated, not only with well known physiological artifacts (blinking, muscle) and line noise etc., but also by CI artifact. The CI induces an artifact in the recording at the electrodes in the temporal lobe area (where it is implanted) when specific tones are presented, this artifact in particular makes the detection and analysis of AEPs much more challenging. This paper evaluates the convenience of using Blind Source Separation (BSS) and Independent Component Analysis (ICA) in order to identify the AEPs from ongoing recordings and to isolate the artifact when testing a child with a CI. We propose a new procedure to elicit an objective differentiation between the independent components (ICs) related to the AEPs and CI artifact; two concepts are fundamental in this procedure Mutual Information (MI) and Clustering. Finally, the variability of three BSS/ICA algorithms is assessed; in order to determine which one is more convenient to isolate the respective ICs of interest. Temporal decorrelation based ICA showed the least change in the estimation of both the AEPs and the CI artifact; this has allowed for considerable autonomy in the construction of relevant, consistent clusters.

Blind source separation by independent component analysis has been applied extensively in the biomedical field for extracting different contributing sources in a signal. Regarding lung sounds analysis to isolate the adventitious sounds from normal breathing sound is relevant. In this work the performance of FastICA, Infomax, JADE and TDSEP algorithms was assessed using different scenarios including simulated fine and coarse crackles embedded in recorded normal breathing sounds. Our results pointed out that Infomax obtained the minimum Amari index (0.10037) and the maximum signal to interference ratio (1.4578e+009). Afterwards, Infomax was applied to 25 channels of recorded normal breathing sound where simulated fine and coarse crackles were added including acoustic propagation effects. A robust blind crackle separation could improve previous results in generating an adventitious acoustic thoracic imaging.

Many authors have used the Auditory Evoked Potential (AEP) recordings to evaluate the performance of their ICA algorithms and have demonstrated that this procedure can remove the typical EEG artifact in these recordings (i.e. blinking, muscle noise, line noise, etc.). However, there is little work in the literature about the optimal parameters, for each of those algorithms, for the estimation of the AEP components to reliably recover both the auditory response and the specific artifacts generated for the normal function of a Cochlear Implant (CI), used for the rehabilitation of deaf people. In this work we determine the optimal parameters of three ICA algorithms, each based on different independence criteria, and assess the resulting estimations of both the auditory response and CI artifact. We show that the algorithm utilizing temporal structure, such as TDSEP-ICA, is better in estimating the components of the auditory response, in recordings contaminated by CI artifacts, than higher order statistics based algorithms.