By Diego Freire Agency FAPESP – Researchers at the School of Medicine (EPM) of the Federal University of São Paulo (UNIFESP) have developed software that detects facial expressions related to the feeling of pain in infants , which may assist in the care of babies, enabling faster and more precise interventions.
The computer program was designed within the software development research to identify the facial expression newborn pain, led by Ruth Ginsburg supported by FAPESP.
According to Ginsburg, the initiative arose from the difficulty faced by caregivers of newborns in intensive care units (ICU) in the recognition and in the assessment of pain signals.
“This subjectivity makes it difficult any interventions, as there are a number of factors that could cause the baby to show some uncomfortable not always related pain. The research enables a useful tool to monitor the baby’s pain during routine neonatal units, “said the Agency FAPESP.
In children who are not yet able to verbalize, recognition of pain is based on behavioral and physiological indicators such as simple motor responses, facial expressions and crying.
The software designed at UNIFESP was developed based on the scale Neonatal Facial Coding System (NFCS), widely used in the recognition of facial movements of pain converted by researchers in computer language with the assistance of the Department of Health Informatics (DIS) of EPM and professionals at the University of Mogi das Cruzes.
The software mechanisms began to be designed in 2009, after approval by the Ethics Committee of UNIFESP. 30 newborns were filmed at the Hospital São Paulo, between June and August 2013.
“very careful work with families so that there were disagreements about the capture of the images was necessary, made during painful procedures with medical indication, such as lancing, venous or arterial and intramuscular or subcutaneous injection, necessary for the care of newborns and unrealized for research purposes, “Ginsburg said.
The selected infants were between 24 and 168 Born hours, without any respiratory support or gavage and without malformations. Those responsible for them signed consent order to be realized catches of the images used in the research.
During the monitoring period, the facial expressions were photographed in real time by three cameras positioned left, right above the newborn.
The software, based on biometric identification, mapped and found 66 points of the face of babies, then reduced to 16 main nodes from which we selected those most moved when it was expressed acute pain caused by some medical procedure
The distances between points served as the basis to detect facial expressions, according to the scale adopted by the survey, show signs of pain:. forehead protruding, narrowed palpebral fissure, depth nasolabial folds, mouth open and tense mouth.
5,644 images have been identified, an average of 188 per infant. Next, the researchers tested the correlation between the analysis software and the six health care professionals experienced in the recognition of neonatal pain, specializing in neonatology. Three images of each baby were compared:. Two registered in the rest period, without pain, and during painful procedures
“We found that the software did not detect expressions of pain in 85% of pictures taken while babies rested without were being subjected to any painful procedure. Already during the procedures, 100% of the images were detected expressions of pain by the program, while some professionals in the identified only 77% of the shots, “said Ginsburg.
Baby sick
To Ginsburg, the precision with which the software detected the expressions provides make it an important tool to assist health teams in neonatal ICUs.
“The newborn’s pain is always seen by others and decoding depends on a caregiver, subject to the time available for monitoring and even more subjective factors, such as empathy. The automation of this monitoring can contribute to the well-being of babies and assist in the care, leading to more assertive interventions, “said
The researchers are now working on adapting the software for monitoring of sick babies, improving the cameras and adjusting the system so that it can be used at the bedside.
Search results were published in article Tatiany Marcondes Heiderich, Ana Teresa Stochero Leslie and Ginsburg, all of the Department of Pediatrics and Applied Sciences the Pediatrics of UNIFESP, in Acta Paediatrica magazine, available in onlinelibrary.wiley.com/doi/10.1111/apa.12861/full.
No comments:
Post a Comment