EXAMINATION OF ELECTRIC FIELD EFFECTS ON LIPID PEROXIDATION AND ANTIOXIDANT ENZYMES BY USING MULTILAYER PERCEPTRON NEURAL NETWORK

Abstract

The aim of this study is to determine lipid peroxidation and antioxidant enzyme levels in liver and lung tissues of guinea pigs which were exposed to different intensities of electric fields and the experimental results are applied to neural networks as learning data and the training of the feed forward neural network is realized. At the end of this training; without applying electric field to tissues, the determination of the effects of the electric field on tissues by using computer is predicted by the neural network. Electric potentials were applied to the copper plates mounted on the wooden boxes to produce electric fields with magnitudes of 1.8 kV/m, 1.35 kV/m, 1 kV/m, 0.8 kV/m and 0.3 kV/m. Male white guinea pigs (150-200 g) were continuously exposed to electric fields for 8 hours per day over 3 days. A total of 100 guinea pigs were exposed to electric fields. Each group of 20 guinea pigs was exposed to the electric field from 9 a.m. to 5 p.m. Twenty guinea pigs were used as controls. The effect of electric field exposure on malondialdehyde (MDA) and superoxide dismutase (SOD) levels was investigated for different intensities. After the experiments, the prediction of the neural network is averagely 97.27% - 99.51 %. Those percentiles of the prediction performance of the neural network belonging to experiment results of electric field were so high; this fact shows that the feed forward neural networks which are used many fields could be applied in the studies of electric field too.