Research summary
Machine learning (ML) has led to significant advances in computational psychiatry and is increasingly used in the field of mental health. But current models often rely on retrospective data from small, selected samples, so are unsuitable for clinical or government use. There is a critical need for ML models that use population-level data to provide individualized predictions of mental and substance use disorder outcomes.
As Canada Research Chair in Computational Psychiatry, Dr. Bo Cao is developing computational models that focus on dynamic predictions in key areas: predicting opioid overdose, detecting depression earlier, and identifying ADHD in children. By collaborating with end users and community stakeholders, he and his research team are validating these models in clinical and governmental settings to ensure they are accurate, reliable and beneficial. This research promises to create practical tools that will ultimately improve the outcomes of patients suffering from mental health and substance use disorders.