Automatic, objective, and efficient measurement of pain using automated face analysis

Abstract

Pain typically is measured by patient self-report, but self-reported pain is difficult to interpret and may be impaired or in some circumstances not possible to obtain. Automatic, objective assessment of pain from video or camera input is emerging as a powerful alternative. We review the current state of the art in automatic, objective assessment of pain from video or camera input and the databases that have made progress in this area possible. Because most efforts have involved facial expression of pain, we emphasize that in our review. We discuss current challenges and prospects to advance automatic assessment of the occurrence and intensity of pain for research and clinical use.