Increasing the Utility of Machine Learning in Clinical Care
Today's hospitals are collecting an immense amount of patient data. My group aims to develop the machine learning tools needed to detect patterns in these data that can help inform clinical decisions, leading to improved patient care. Through numerous collaborations with clinicians, we have identified key characteristics necessary for the safe and meaningful adoption of clinical decision support tools. In addition to being accurate, risk stratification models must be actionable, credible, and robust to changes over time. To achieve these goals, we are investigating novel ways of incorporating clinical domain expertise during model training and development. In this talk, I will present ongoing work focused on extensions to deep learning architectures and regularization techniques that explicitly leverage domain expertise. I will show that through the thoughtful incorporation of existing expert knowledge, we can increase the utility of such models, even in the high-dimensional low-sample size setting that is common in this field.
Jenna Wiens is an Assistant Professor of Computer Science and Engineering (CSE) at the University of Michigan in Ann Arbor. Her primary research interests lie at the intersection of machine learning, data mining, and healthcare. She is particularly interested in time-series analysis and transfer/multitask learning. The overarching goal of her research agenda is to develop the computational methods needed to help organize, process, and transform patient data into actionable knowledge. Jenna received her PhD from MIT in 2014. In 2015 she was named Forbes 30 under 30 in Science and Healthcare; she received an NSF CAREER Award in 2016; and recently she was named to the MIT Tech Review's list of Innovators Under 35.