Denoising of Transient Visual Evoked Potential using Wavelets

Abstract

Transient Visual Evoked Potential (TVEP) is an important diagnostic test for specific ophthalmological and neurological disorders. The clinical use of VEP is based on the amplitudes and the latencies of the N75, P100, and N145 peaks. The amplitude and latencies of these peaks are measured directly from the signal. Quantification of these latency changes can contribute to the detection of possible abnormalities. We have applied the wavelet denoising method to 100 numbers of pre-recorded signals using all the available wavelets in MATLAB signal processing toolbox. From the results it is clear that the positive peak is clearer in the denoised version of the signal using the wavelets Sym5 and Bior3.5. As opposed to the previous studies however our study clearly shows that the output using the former wavelet effectively brings out the P100 when compared to the latter. The first negative peak N75 is clear in the denoised version of the signal using the wavelets Bior5.5 Bior6.8 and coif4. The second negative peak N145 is clear using all the above wavelets. All the three peaks are fairly clear in the denoised output using the wavelet sym7.