Camera-based Pain Assessment during Surgical Interventions

Abstract Pain assessment in clinical settings can be a challenging task for clinic staff and patients. An automatic contactless pain estimation system via facial expression is decisive advantage. In the following article we present method classification based on mimic descriptors. To automatically process expressions from video data applied Facial Action Coding System (FACS). First, employed logistic regression model to predict using publicly available database. Second, utilized this procedure individual intraprocedural patient data. We reached mean accuracy of 80 . 94% 73 14% our by classifying as non-pain.