My research focuses on the reuse of routinelly collected clinical data for research. This has both theoretical and applied aspects.

On the theoretical side, I work on developing and testing algorithms and query methods to identify patient cohorts based on temporal patterns of events that can be found in EHR databases. I call this Temporal Abstraction-Based Digital Phenotyping. You can find some of our results here, and here. I am currently applying these methods to quickly replicate randomized trials using digital phenotypes and real-world data and identify sources of limited external validity.

On the applied side, I work with a multidisciplinary research team from the School of Engineering where we apply Healthcare Process Mining techniques to better understand clinical processess and how they may impact patient outcomes. Some publications related to this can be found here, here, here, and here. We recently combined Temporal Abstraction-Based Digital Phenotyping with process mining to discover drug use patterns in the ICU for patients with and without Sepsis.

I am also on the board of directors of the Epistemonikos Foundation, the worlds largest database of verified systematic reviews. The foundation focuses on developing informatics tools to quickly visualize clinical evidence, create clinical guidelines and systematic reviews that are able to automatically update themselves.

© Daniel Capurro 2018