A deep neural network-based pain classifier using a photoplethysmography signal

Abstract

Side effects occur when excessive or low doses of analgesics are administered compared to the required amount to mediate thepaininduced during surgery. It is important to accurately assess thepainlevel of the patient during surgery. We proposed apainclassifierbased on adeepbelief network (DBN) usingphotoplethysmography(PPG). Our DBN learned about a complex nonlinear relationship between extracted PPG features andpainstatus based on the numeric rating scale (NRS). A bagging ensemble model was used to improve classification performance. The DBNclassifiershowed better classification results than multilayer perceptronneuralnetwork (MLPNN) and support vector machine (SVM) models. In addition, the classification performance was improved when the selective bagging model was applied compared with the use of each single modelclassifier. Thepainclassifierbased on DBN using a selective bagging model can be helpful in developing apainclassification system.