Guest Speaker: James Jones, MD, MSc, FASA

Institution: UNC School of Medicine

Objectives: 

By the end of this presentation, participants will be able to:

1. Describe the historical evolution of preoperative optimization, from early risk stratification models to modern data-driven perioperative care.
2. Explain the principles and assumptions of logistic regression as the traditional foundation of perioperative risk prediction.
3. Compare logistic regression with contemporary machine learning approaches, including support vector machines, random forests, and gradient boosting models.
4. Interpret key differences between statistical inference and predictive modeling in the context of perioperative decision-making.
5. Discuss how machine learning can enhance preoperative optimization by identifying modifiable risk factors and enabling individualized perioperative interventions.
6. Identify practical opportunities and limitations for implementing machine learning tools in clinical preoperative workflows.