This facial recognition software evaluates pain in children – unocero

To assess accurately the pain of a child in a clinical setting can be tricky ; that is why researchers at the University of California San Diego (UCSD) have developed a facial recognition software that has the ability to assess the level of pain in hospitalized children.

Currently, to determine the pain , doctors ask hospitalized children to rate the pain on a scale of 0-10, but this method can not sufficiently reliable and less with children who are too young and not yet well understood numerical system.

The researchers report that used the software to analyze facial expressions related to pain registered in videos made 50 young people, aged between five and 18, who had undergone laparoscopic appendectomies. Based on this analysis, along with clinical data, the software gave a score to each patient

In the study, researchers recorded the participants in three different visits after surgery. 24 hours the appendectomy , a day after the first visit and a follow-up of 2-4 weeks after surgery. Expressions are collected much (considering 46 movements of facial muscles) and self-reported pain scores by participants as the qualifications of parents and nurses.

“Software demonstrated an accuracy of” good to excellent “in assessing pain conditions. Overall, the results were equivalent to those estimates parents and better than those of nurses, with a strong correlation to the report of the patients themselves. In addition, the software did not detect any bias in the evaluation of pain by ethnicity, race, sex or age among the patients studied, “said Jeannie Huang, head of research.

In addition, developers stand its usefulness in the absence of the parents , since when are unavailable to notify the medical staff about the level of pain of their children, as well as the ability to continuously monitor the degree of pain, which is not can be performed with assessments of pain that can make parents or medical personnel and are held at scheduled times.

Now the goal of researchers is to expand the study to other types of clinical pain and full range of childhood. “There’s still that determine whether a tool of this kind can be easily integrated into the clinical workflow and thus bring benefits to both current assessment methods as the treatments themselves,” he concluded.

Reference: UCSD