Machine Learning for Patient Risk Stratification

Chapters 1 and 2 were foundational chapters that were designed to help you get introduced to the concepts of ML, its applications in healthcare and life sciences, and also some key services from AWS. With this foundational knowledge, we can now start applying these techniques to real-world industry problems. Over the course of the next eight chapters, you will learn about specific applications of AI/ML for solving problems for healthcare providers, payers, pharma, medical devices, and genomics customers.

In this chapter, we will look at one of the most common usages of ML in healthcare, the stratification or identification of risky patients. We will learn what it takes to identify risky patients and the common ML models that can help with this identification. We will then implement an example ML model that identifies patients at risk of breast cancer using SageMaker Canvas, the low-/no-code service from AWS that allows citizen data...