It's important to know how much pain young hospital patients are experiencing, and not just because no one wants them to suffer – additionally, excessive pain can indicate problems that need addressing. That's why scientists at the University of California, San Diego School of Medicine have developed facial pattern recognition software that objectively assesses children's pain levels based on consistent indicators.
Traditionally, caregivers ask children to rate their pain on a level of 0 to 10. This can be a problem with patients who aren't coherent, however, and is particularly challenging with younger children who can't yet grasp the concept of numerical ratings systems.
Parents or nurses are often asked to provide an estimate of the child's pain level in such situations, although the accuracy of such an approach is certainly far from perfect. Additionally, parents may not be on hand when an assessment is needed.
In order to address such limitations, the UCSD team developed software based on the Facial Action Coding System (FACS). Utilizing a video camera trained on the patient's face, it tracks 46 different facial movements, and uses them to assign a real-time "pain score."
In a test of the system, it was used on 50 laparoscopic appendectomy patients aged 5 to 18 years old. Their faces were recorded in sessions that took place 24 hours after surgery, one day after that first visit, and then again two to four weeks later. During those sessions, the patients provided their own self-reported pain ratings, plus estimates were obtained from parents and nurses.
"The software demonstrated good-to-excellent accuracy in assessing pain conditions," says Dr. Jeannie Huang, senior author of a paper on the research. "Overall, this technology performed equivalent to parents and better than nurses. It also showed strong correlations with patient self-reported pain ratings."
Factors such as ethnicity, race, gender and age appeared not to affect the accuracy of the software.
The system also has another advantage over traditional methods, in that it could be used continuously. Whereas pain assessments are usually performed at scheduled intervals, it's possible that a patient might experience their most painful episodes between those times. If the software was monitoring them all the time, it could record the severity of that pain, and alert caregivers to provide intervention.
The research paper was published this week in the journal Pediatrics.