We now have available ever-increasing amounts and types of data related to health: personal data on weight, activity, diet, sleep and heart rate collected from wearable devices, home technologies and smartphone apps; electronic medical record data including genomic data, biomarkers, blood pressure, glucose, cholesterol and a host of others; data on environmental and social influences important to health such as air quality, noise, exposures to walkable communities and green space or to incivilities, and data on social networks. Some of these data are streaming and ever-changing while others are measured and updated periodically. Much of it is siloed in proprietary “walled gardens” that limit its usefulness to the very people it could benefit. Still others are only important if they can be used in a comparative way or referenced to a rapidly changing landscape of medical, public health and scientific knowledge. Moreover, the tremendous volume and heterogeneity of these data makes drawing inferences from them for any one person a daunting, if not impossible task.

How can we empower patients, providers, consumers and app developers with access to these data so that they can help us move to personalized population health? With the advances in databases and machine learning proposed for the Data E-platform Leveraged for Patient Empowerment and Population Health Improvement (DELPHI), a new category of healthcare applications that infer one’s health status in the perspective of one’s entire life history and context is anticipated.