WAVELET ANALYSIS OF HUMAN PHOTORECEPTORAL RESPONSE

Feature detection of biomedical signals is crucial for deepening our knowledge of the physiological phenomena giving rise to them. To achieve this aim, even if many analytic approaches have been suggested only few are able to deal with signals whose features are time dependent, and to provide useful clinical information. In this work we use the wavelet analysis to extract peculiarities of the early response of the photoreceptoral human system, known as a-wave ERG-component. The analysis of the a-wave features is important since this component reflects the functional integrity of the two populations of photoreceptors, rods and cones whose activation dynamics are not well known. Moreover, in incipient photoreceptoral pathologies the eventual anomalies in a-wave are not always detectable with a naked eye analysis of the traces. We here propose the possibility to discriminate the pathologic from the healthy traces throughout the differentiation of their time-frequency characteristics, revealed by the wavelet analysis. The investigated pathologies are the Achromatopsia, a cone disease and the Congenital Stationary Night Blindness, a rod trouble. The results show that the number of stable frequencies present and their times of occurrence are indicative of the status of the retinal photoreceptors. In particular, in the pathological cases, the frequency components shift toward lower values and change their times of occurrence, with respect to healthy traces.