Artificial neural networks: applications in pain physiology

Artificial neural networks (ANNs) are intelligent systems that have successfully been used for prediction in different medical fields. In this study, the capability of ANN in predicting body behavior in pain-producing situations is evaluated. A three-layer back-propagation ANN is designed using MATLAB software. The inputs include the magnitude of stimulation in pain fibers, touch fibers and central anti-nociceptive fibers and output is the level of perceived pain. In other words, we modeled the gate control theory of pain. Important features of pain process were chosen and defined in 8 features and then were applied to the ANN. We examined the ANN to ensure that it can model the real situations. The result was acceptable (errors below 1%). Our model can be used for interpolation and extrapolation of pain-related data. This model is a useful tool in pain experiments to predict the behavior of the organism.