Computational algorithms that can process large amounts of data are increasingly being applied to predict health outcomes. Machine learning models in particular are emerging as powerful artificial intelligence tools that can be continuously and incrementally improved as data accrues to support medical diagnostics and clinical decision-making. Existing health information systems in Ontario can be leveraged to develop, test and validate prediction algorithms and machine learning models in maternal, newborn and child health specialties. 

This research program involves applying machine learning to large clinical and administrative datasets to create prediction models capable of identifying populations at high-risk of adverse obstetrical, infant and child health outcomes, and those most likely to benefit from specialized care and pharmacological and surgical interventions. Findings from this work will help identify gaps in care, target treatments and interventions to those most likely to benefit and improve outcomes. 

• Improving the health of infants and children through research that links large health administrative databases with clinical, screening, and laboratory data. Canadian Institutes for Health Research. Catalyst Grant: Personalize Health Catalyst Grants. Principal Applicants: Hawken S, McNally JD. ($197,610 CAD) (2017) 
  
• The missing link: Integrating placental pathology into existing pregnancy, birth and childhood health repositories. University of Ottawa Faculty of Medicine Translational Research Grant. ($37,500) Principal Applicants: Hawken S, Bainbridge S, El Demellawy D, Grynspan D. (2018) 
• Harnessing the power of artificial intelligence for prediction of an expectant mother’s real-time risk of preterm delivery. Health Research Foundation. Artificial Intelligence Health Research Fellowship Grant ($100,000) Principal Applicant: Hawken S. (2019)

• Walker MC, Willner I, Miguel O, Murphy MSQ, El-Chaar D, Moretti F, Dingwall Harvey ALJ, Rennicks White R, Muldoon KA, Carrington AM, Hawken S, Aviv RI. Using Deep-learning Algorithms in Fetal Ultrasound Analysis for Diagnosis of Cystic Hygroma in the First Trimester. PLOS ONE 17(6): e0269323. doi: 10.1371/journal.pone.0269323 
• Hawken S, Murphy MSQ, Ducharme R, Bota AB, Wilson L, Cheng W, Tumulak MJ, Alcausin MML, Reyes MA, Qiu W, Potter BK, Little J, Walker M, Zhang L, Padilla C, Chakraborty P, Wilson K. External validation of machine learning models including newborn metabolomic markers for postnatal gestational age estimation in East and South-East Asian infants. version 2; peer review: 1 approved, 3 approved with reservations]. Gates Open Research 2021, 2021, 4:164. doi:10.12688/gatesopenres.13131.1 
  
• Murphy MSQ, Hawken S, Cheng W, Wilson L, Lamoureux M, Henderson M, Pervin, J, Chowdhury AKA, Gravett C, Lackritz E, Potter B, Walker M, Little J, Rahman R, Chakraborty P, Wilson K. Postnatal gestational age estimation using newborn metabolic profiles: A validation study in Matlab, Bangladesh. eLife. 2019; 8: e42627. doi: 10.7554/eLife.42627. 
  
• Wilson K, Hawken S, Murphy MSQ, Atkinson KM, Potter B, Sprague A, Walker M, Chakraborty P, Little J. Postnatal prediction of gestational age using newborn fetal hemoglobin levels. EbioMedicine. 2017; 15:203-209. doi: 10.1016/j.ebiom.2016.11.032.