This paper takes a data mining approach to study adherence to continuous glucose monitors in diabetes management.

In this work, we conducted a two-phase user-study involving patients, caregivers, and clinicians to understand gaps in current approaches that support reflection and user needs for new solutions.

In this paper, we employ a data-driven approach to study the relationship between key behavioral factors (sleep, meal size, insulin dose) and proximal diabetic management indicators.

In this paper, we present Neural Physiological Encoder (NPE), a simple module that leverages decomposed convolutional filters to automatically generate effective features that can be used with a downstream neural network for blood glucose prediction.