What does it mean to be healthy? What causes variations in phenotypes (our observable traits, like height, eye color or blood type)? A greater understanding of these questions could enable scientists to improve health care and quantify the impact of possible interventions.
To that end, Dr. Marzyeh Ghassemi, Canada Research Chair in Machine Learning for Health, is exploring representation learning (a form of machine learning) of phenotypes. Because different data may be important for a particular patient or task, Ghassemi and her team are integrating notes, procedures, diagnoses, vitals, labs and other signals in their work. They hope their research will have an impact on our understanding of how human phenotypes can vary and what it means to be healthy.