Explicit Formulation of Magnetic Fields Effects on Skin Collagen Synthesis Via Neural Networks

Abstract

It is well known that the electricity and electrical equipment which emits electromagnetic fields (EMFs) are used increasingly.Thus many people are subjected to environmental extremely-low-frequency -especially 50-60 Hz- electromagnetic fields. This study focused on two purposes: First one was planned to observe experimentally whether the skin of guinea pigs was affected by magnetic fields by determining the collagen synthesis in the skin exposed to 50 Hz magnetic fields of 1 mT, 2 mT and 3 mT with the periods of 4 hours/day and 8 hours/day for 5 days which were determined by Woessner’s method. Secondly, it was aimed to get analytical expression to model the ELF effect on hydroxyproline concentrations in the skin using Neural Networks (NNs). One of the important tasks regarding these types of studies, there will be no need to extent the experiments for different range of EMFs exposing animals although the experimental results serve as a database for researchers. The experimental results as learning data and the training of the feed forward NN are applied to neural networks by using a novel approach for the selection of optimal NN architecture. The accuracy of the chosen optimal NN model is defined by standard deviation STD=0.03 and correlation coefficient R=0.98 which is found to be quite high.Thus parametric studies as a formulation are performed to see the influence of each parameter by using the proposed NN model. At the end, this NN training causes to determine the EMFs effect on tissues without exposing tissues to EMFs and without using too many guinea pigs