Data types and resources

Abstract Recent innovation in the field of machine learning has been enabled by the confluence of three advances: rapid expansion of affordable computing power in the form of cloud computing environments, the accelerating pace of infrastructure associated with large-scale data collection and rapid methodological advancements, particularly neural network architecture improvements. Development and adoption of these advances have lagged in the health care domain largely due to restrictions around public use of data and siloed nature of these datasets with respect to providers, payers and clinical trial sponsors.