Introduction and Course Overview

Machine learning (ML) and artificial intelligence (AI) methods are expected to have a transformative impact on innovation and discovery in healthcare. While ML & AI methods have been around for nearly 60 years, a combination of recent advances in methods, access to large amounts of data, and powerful computing have made it possible to bring them to bear on various problems in healthcare. Together with commitment from funding and regulatory agencies, the time is now ripe for AI to make its mark in healthcare by solving pressing problems using data.

ML & AI methods are now rapidly becoming a staple in healthcare research. Throughout history, methods from the quantitative disciplines have served to improve healthcare, either through advanced measurements, analytics, or inference. For example, methods from statistics, education, psychology, and epidemiology have enriched healthcare through research that was previously not possible. Our society is now at such a juncture with ML & AI methods.

The key to realize the potential for ML & AI in healthcare is trained stakeholders, including not only care providers and decision-makers but also engineers and data scientists. ML & AI in healthcare is a team science and it is now considered a new “fundamental competency for healthcare providers” (1). To enable ML & AI in healthcare, we must bridge the knowledge gap between healthcare decision-makers and engineers/data scientists. Specifically, all stakeholders should understand ML & AI methods and how algorithms are validated for healthcare applications.

This course is being developed with funding from a Digital Education & Learning Technology Acceleration (DELTA) grant from Office of the Provost, Johns Hopkins University.

1. Reznick, et. al. Task Force Report on Artificial Intelligence and Emerging Digital Technologies. Royal College of Physicians and Surgeons of Canada.