Katharine Henry

I am a postdoc in the computer science department and the Malone Center for Engineering in Healthcare at Johns Hopkins University. I work with Suchi Saria in the Machine Learning and Healthcare Lab, where I also completed my PhD in 2020 and was an NSF graduate research fellow.

My research primarily focuses on how machine learning (ML) can be used to improve healthcare. Using electronic health record (EHR), I focus on develping solutions to messy real-world clinical problems including applications like developing acute deterioration monitoring tools and using causal inference to improve the reliability of predictions. My research also explores how clinicians interact with and perceive ML-based clinical support systems and uses human factors appraoches in order to develop strategies to improve human and machine collaboration in the clinical space.

During my PhD I developed and deployed the Targeted Real-time Early Warning System for Sepsis (TREWS), a real-time machine learning alert for sepsis that is embedded into the EHR at five hospitals, as part of a collaborative team of engineers, clinicians, and patient quality and safety experts. My disseration titled Translating Machine Learning into Clinical Practice: Lessons from Development to Deployment is based on this work.