Evaluation of the effectiveness of the decision support algorithm for physicians in retinal dystrophy using machine learning methods

Electroretinography is a method of electrophysiological testing, which allows diagnosing diseases associated with disorders the vascular structures retina. The classical analysis electroretinogram based on assessing four parameters amplitude-time representation and often needs to be specified further using alternative diagnostic methods. This study proposes use an original physician decision support algorithm for retinal dystrophy. proposed machine learning methods uses extracted from wavelet scalogram pediatric adult signals. also labeled database signals recorded computerized workstation EP-1000 (Tomey GmbH) at IRTC Eye Microsurgery Ekaterinburg Center. scientific novelty this consists in development special mathematical algorithmic software analyzing procedure extracting signal cwt function PyWT. basis Gaussian order 8. Also, includes that implements classification (pediatric) 19 (20) percent more accurately than analysis.