The healthcare industry has yet to fully embrace machine learning and all it has to offer. But why is that? This session will examine some of the ways machine learning is currently used in medicine and challenges that the industry has faced, uncovering some success stories and areas for improvement. Key takeaways: The healthcare industry needs to shift from using data reactively to a more measured, proactive approach. To get the best results when formulating a machine-learned tool: Start with a precisely defined performance task, not a dataset. Use relevant data from appropriate populations of patients.