Advanced signal processing techniques for trigeminal somatosensory evoked potentials

Very weak and noisy trigeminal somatosensory evoked potentials (TSEPs) are considered, which were successfully evoked by electrical stimulation of the trigeminal nerve of 15 patients with endosseous oral implants. As TSEP testing may offer an objective means of assessing neuronal function, the aim was to estimate the tactile sensation through the anchoring implant in the bone. For this purpose, a challenging study of TSEP signals from patients with endosseous oral implants was carried out. Since TSEPs are severely contaminated by background ongoing electrical activities of the brain, a methodology was designed for statistically detecting transient signals (TSEPs) in electroencephalograms (EEGs). For 9 out of 15 patients, we were able to detect the transient signal in the background EEG activity. As a consequence, the TSEPs of these 9 patients were used for further analysis. Subsequently, a multi-channel singular value decomposition (SVD) based filtering method is presented, which is a new and promising technique for the adaptive filtering of TSEPs, thereby improving the SNR of the recorded signals. Multichannel SVD-based filtering is an appropriate noise reduction method which enables the extraction of the characteristic components of the TSEPs. We have found that the most prominent feature of the TSEP signals for patients with endosseous oral implants is a wave with a peak latency between 9-15 ms, which is mostly followed by a wave between 25-28 ms or 34-38 ms for the specific cortical response areas.