Grant: Doctoral - PA - Canada Graduate Scholarship to Honor Nelson Mandela

Every year, about 8% of Canadian babies are born prematurely. Many of these infants require care in the Neonatal Intensive Care Unit (NICU), where they are exposed to numerous painful procedures. Early exposure to pain can result in long-term changes to brain structure and the stress response system, leading to both immediate and long-term developmental challenges. Currently, pain assessment in NICU infants relies solely on healthcare professionals' interpretation of behavioural cues (e.g., facial expressions, crying) and physical measures (e.g., heart rate, oxygen saturation). However, these methods are prone to caregiver judgment bias, lack specificity in distinguishing pain from other forms of distress, and are challenging to apply consistently in busy hospital environments (Pillai Riddell, Jasim & Hamwi, 2022). Building on my Master's thesis, this proposed study aims to use machine learning to develop a more accurate way of detecting pain responses in preterm infants. By analyzing data from brain activity (EEG), heart rate, breathing, and facial behaviour, we will create a computer algorithm capable of quickly and accurately distinguishing between pain responses following a painful procedure (i.e., heel lance), a resting state, and a non-painful matched procedure (i.e., sham lance with no skin break). This technology will assist healthcare professionals in managing pain more effectively for these vulnerable patients, improving care and outcomes. In the long term, this project could revolutionize how pain is monitored in NICUs, offering real-time assessments that enable doctors and nurses to respond more effectively to infant pain. Ultimately, this could lead to healthier development for preterm infants and enhance the quality of care in hospitals across Canada and beyond.