On April 2, a team of York University researchers led by psychology Professor Rebecca Pillai Riddell, associate vice-president research and the director of the Opportunities to Understand Childhood Hurt (O.U.C.H.) Lab, was awarded a $1.5-million grant from the Canadian Institutes of Health Research, the Social Sciences & Humanities Research Council, and the Natural Sciences & Engineering Research Council of Canada – plus $387,000 in-kind contributions from community partners.
The project being funded, “Rebooting Infant Pain Assessment,” could give voice to preterm infants’ subjective experiences of pain, despite their inability to speak. The study will use machine learning to exponentially improve neonatal intensive care unit practice.
Pillai Riddell is the principal investigator (PI), with Professors Steven Wang (co-PI, Faculty of Science) and Aijun An (co-investigator, Lassonde School of Engineering) and Ian Stedman (Osgoode Hall Law School). Pillai Riddell is leading a team of 16 individuals across two countries and six sites. This is a highly collaborative international venture and it exemplifies cross-Faculty collaboration at York University.
“The AI (artificial intelligence) component in this project is important because it is enabling us to use continuous brain activity in a proposed clinical pain assessment application – to our knowledge, that’s a first anywhere in the world,” says Pillai Riddell. “York’s partnership with UCL (University College London) and McMaster University in this venture is a fantastic synergy of our combined expertise of behavioural and biological infant distress responses. I am thrilled to then be able to take our health content knowledge and take it to the next level with cutting-edge York U artificial intelligence scientists in two sector-leading neonatal intensive care units – one in Canada (Mount Sinai) and one in the U.K. (University College London Hospital).
“Moreover, this special Tri-Council opportunity inspired us to invite new social scientists at Osgoode Hall Law School (Ian Stedman) and University of Calgary to explore the ethical and social implications of computer-assisted clinical decision-making,” she adds.
Desperate need for a better way forward for infant pain assessment
The need is great. Unmanaged pain in hospitalized infants has serious long-term complications. However, to manage pain, one must have accurate infant pain assessment. Infants cannot self-report their pain and current infant pain assessment tools used by health professionals have major problems because of the lack of specificity of current tools and bias in the caregivers who use these scales.
The researchers believe they have found a path towards a solution. “Our international team of knowledge users and health/natural science/engineering/social science researchers have come together to build a machine learning algorithm that will learn how to discriminate invasive and non-invasive distress,” Pillai Riddell explains.
Three hundred babies and their mothers will be studied
A sample of 300 preterm infants and their mothers will be involved during a routine painful procedure. Pain indicators, such as facial grimacing, heart rate, brain electrical activity and oxygen levels will be used to train the algorithm to discriminate between the different types of distress.
“The complexity of pain requires a machine learning solution that is capable of modelling individual patterns of brain, behaviour and physiology during pain,” Pillai Riddell explains.
This article was originally posted on yFile.
