Advanced Analysis of PRVEP in Anisometropic Amblyopia

to identify the pattern-reversal visual evoked potential (PRVEP) waveform descriptor by evaluating discrete wavelet transform (DWT) in order to optimize stimulus in the diagnosis of anisometropia amblyopia.
The PRVEP testing was performedfor 31 normal individuals and 35 patients with amblyopia. The stimuli were consisted of spatial frequencies of 1, 2, and 4 cycles per degree (cpd) and contrast levels of 100%, 50%, 25%, and 5%. The results were analyzed in the dimensions of time and time-frequency. DWT descriptor were extracted at level 7 (7P descriptor) for Haar, Daubechies 2, Daubechies 4, Symlet 5, Biorthogonal 3.5, Biorthogonal 4.4, and Coiflet 5 wavelets for 12 stimuli and compared between the two groups. The correlation between different spatial frequencies at the same contrast level and the similarities between reconstructed signals and original waveforms were evaluated.
There were a significant reduction in P100 amplitude and a significant elevation in latency among the patient group. In the patients with amblyopia, 7P descriptor decreased in all analysis except for the frequency of 4 cpd and the contrast of 5% using bior4.4. No significant correlation was observed between different frequencies at a special contrast; however, there was a significant correlation between reconstructed signals and the original ones.
The 7P descriptor could be used to distinguish between normal and abnormal signals in anisometropia amblyopia. Considering the results, DWT with coif5, db4, bior4.4, and bior3.5 wavelets can be utilized as a good indicator for selecting optimum stimulus.