With all the buzz around machine learning, predictive analytics, and artificial intelligence (AI) there are a lot of misconceptions and misunderstandings surrounding the optimal use of modern machine learning tools. Healthcare.ai, a free software package developed by the Health Catalyst data science team, was recently released to help hospitals gain valuable insights and advance outcomes improvements from their immense data sets. The software automates machine learning tasks and democratizes machine learning by making it accessible to ‘citizen data scientists’. We have received several questions about machine learning in healthcare, such as how do you define machine learning, how is it different than AI, what are some common uses cases for machine learning in healthcare, and what are the pitfalls. This webinar will develop a common vocabulary around these ideas. We’ll cover the differences between the most cutting-edge predictive techniques, how a model can be improved over time, and use case vignettes to understand and avoid typical machine learning pitfalls. In today’s healthcare industry, the fastest path to healthcare outcomes is often achieved using the simplest predictive tools.
Mike Mastanduno, PhD, data scientist, and Levi Thatcher, PhD, director of data science, will discuss the landscape of healthcare-specific machine learning. Mike and Levi have extensive experience building and deploying impactful machine learning models using healthcare.ai and have worked at the cutting edge of medical research. During and after the discussion, they will answer viewer-submitted questions. This webinar will:
Compare and contrast machine learning and AI.
Discuss techniques that offer feedback into the system and when it’s necessary to retrain a model.
Give advice on how to avoid common pitfalls in machine learning implementation.
Provide use case example and vignette examples on how to apply the different classes of machine learning techniques.
Whether you are on the go or have time for a deep dive, we offer multiple ways to view the content.