Image Registration for Detection and Modeling of P-100 Latency and Amplitude Variability in Multi Focal VEP

Abstract

The analysis of P-100 latency and amplitude measurement variability based on visual pathway conduction in multi focal VEP (Visual Evoked Potential) has been shown to have clinical utility. The reliable measurement of VEP techniques to do so are less well developed. This work presents a technique for the reliable extraction P-100 latency and amplitude using a SVD algorithm. For our preliminary high-quality multi focal VEP recordings of P-100 latency and amplitude measurement methods , this technique is 99.93% effective in reliably detecting measurement. The physiological basis for P100 variability in humans can be examined using these techniques. The challenge of image registration (the process of correctly aligning two or more images accounting for all possible sources of distortion) is of general interest in image processing. This registration can be carried out for images of the same subject taken at different times, images taken under different modalities, and images which have only a partial overlap area. The technique has been applied to different types of retinal images and the technique can be readily used to provide cross-modal registration.