A dynamic prediction model for intraoperative somatosensory evoked potential monitoring

Abstract

This study proposed a support vector regression model applied in prediction of intraoperative somatosensory evoked potential changes associated with physiological and anesthetic changes. This model was developed from probability distribution and support vector machines. The predicted results showed that observed and predicted SEP has similar variation trend with different values, with acceptable errors. With this prediction model, changes of SEP in correlation with non-surgical factors were estimated. Not only the prediction accuracy of SEP has been improved, but also provides the reliability of the classification. It will be helpful to develop an intelligent monitor model based expert system that can make a reliable decision for the potential spinal injury.