Statistical Analysis And Detection Of Diabetes Using Electro-Oculogram
Abstract
Diabetes is a common condition that is predominant in India which occurs due to the increase in the blood glucose level due to certain metabolic abnormalities of the body. Diabetes induces muscle fatigue in all parts of the body including the eye leading distraction in vision. In this study, Electro-oculogram (EOG) can be used to analyse the fatigue of eye by measuring the potential difference between the retina and cornea with respect to the change in position of eye ball .This study aims at analyzing and predicting diabetes by obtaining Electro-oculogram bio-signals. EOG signal is acquired for both diabetic and non-diabetic patient at specific age groups. Two different eye movements are recorded using simple pendulum movement and tracking an object method. Power spectral density used for analyzing the changes in the EOG signal from initial to end to define the peak fatigue level. Statistical features are extracted from the spectral signals for determination of muscle fatigue level. Further, statistical analysis is performed to illustrate the significant nature of the features. With these results, the fatigue in the eye is illustrated that helps in identification and prediction of diabetes.