Presenter: Shriti Raj

Description: Type 1 diabetes (T1D) technology has seen tremendous innovation ranging from insulin pumps and continuous glucose monitors to the artificial pancreas systems. Despite the use of sophisticated technology, few people with T1D meet standard care goals. Among the people using such technology, a key moderator of glycemic outcomes is periodic engagement with self-tracked data to inform care decisions. Existing data reporting tools offer poor support for simultaneously interpreting multiple data streams, resulting in missed opportunities for care. How can we create informatics solutions that enable careful yet low burden translation of multiple streams of self-monitored data to actionable information? In this talk, I will present a series of studies that a) examine the current use of T1D data in and outside the clinic and b) prototype and evaluate human-machine artifact configurations that can better support users in generating value from data. Taken together, these studies highlight the potential of task-driven conceptualization, design, and evaluation of decision-support data interfaces and the need to expand the role of algorithmic support in helping make data actionable.

Bio: Shriti Raj is an Assistant Professor (Research) in Stanford’s Center for Biomedical Informatics Research and a Fellow at the Institute for Human-Centered AI. Her research focuses on developing and evaluating human-centered techniques to help people (patients and clinicians) make health data and algorithms actionable. Shriti received her PhD from the University of Michigan. Before that, she completed a master’s from UC Irvine, a bachelor’s from IIT (India), and worked at Goldman Sachs to create market data platforms.

Hosted by: Professor Christina Chung