Objective Method for Pain Detection/Diagnosis

Multiple studies have identified physiological and behavioral variables that are associated with pain intensity in critical care patients. In this phase of the research, we propose to investigate the feasibility of developing a multi-modality pain intensity detection algorithm predicated on physiological and behavioral indicators of pain as well as designing a plan to"calibrate"and validate the pain score provided by the algorithm. Specifically, we will use the physiological data generated by patient monitoring devices currently in use in the critical care environment as well as introduce the use of a number of sensors in order to complement the information provided by existing sensors. We propose to use machine learning to fuse the multi-modal sensor data and provide an objective assessment of pain intensity. Our proposed approach is adaptive in nature, and hence, can address challenges of intrapatient and interpatient analgesic state variability. In addition, the proposed sensor fusion framework is robust to sensor failure and incomplete sensor measurements making it especially appealing for combat critical care and combat evacuation. The emphasis of the approach will be on a portable objective pain detection device which can be utilized for monitoring nonverbal patients in the combat environment as well as in evacuation scenarios.