Home page

The United States' medical billing system is exceptionally complex. Medical bills and Explanation of Benefits (EOB) statements are undecipherable and incomprehensible even for experts to understand. In addition, a 2015 survey conducted by TransUnion Healthcare found that 55% of American patients were either sometimes or always confused about their medical bills and that 61% of patients were either sometimes or always surprised about their out-of-pocket costs.

Analysts routinely encounter large collections of text documents with many accompanying numeric or categorical data fields. For example, consider a collection of wine reviews where the main text part of the document is the actual review narrative, but the accompanying fields are data such as the wine's variety, color, age, rating, producer, region, reviewer, and so on. In this project, we are exploring techniques to allow analysts to investigate the document collection by "slicing" it along the different attributes affiliated with each document (we call them "facets").

Although it is well-established that the environment (e.g., sound, light, temperature) can be a significant contributor to sleep disruption, few technologies have been designed to capture the environmental factors that affect sleep. We developed a low-cost and robust sensor designed to collect highly accurate environmental light and sound data in a natural home environment at fine temporal resolution. We deploy these sensors in the bedrooms of older adults for an extended period to comprehensively describe the sleep environment and to uncover the metrics most salient for sleep disruption.