Pain Detection Using Deep Learning Method from 3D Facial Expression and Movement of Motion

Abstract

Nowadays, face expression technology is widespread. For instance, 2D pain detection is utilized in hospitals; nevertheless, it has some disadvantages that should be considered. Our goal was to design a 3D pain detection system that anybody may use before coming to the hospital, supporting all orientations. We utilized a dataset from the University of Northern British Columbia (UNBC) as a training set in this study. Pain is classified as not hurting, becoming painful, and painful in our system. The system’s effectiveness was established by comparing its results to those of a highly trained medical and two-dimensional pain identification. To conclude, our study has developed an uncomplicated, cost-effective, and easy to comprehend alternative tool for screening for pain before admission for the public in general and health provider.