Critical Care Data Analytics

Critical care medicine is a data-intensive medical specialty that deals with a large amount of heterogeneous data on a daily basis. Although the volume and variety of critical care data pose many challenges, they also present interesting opportunities for health data science research. Our over-arching objective is to uncover useful information and knowledge from critical care data by applying advanced data analytics.

We primarily utilize the public critical care database called MIMIC which contains rich clinical data from approximately 60,000 intensive care unit admissions at Beth Israel Deaconess Medical Center in Boston, Massachusetts, USA. MIMIC was created and is maintained by the Laboratory for Computational Physiology at MIT. We have used MIMIC data to study hypotension (Lee et al., 2012), acute kidney injury (Fuchs et al., 2013Mandelbaum et al., 2013Mandelbaum et al., 2011), novel mortality predictors (Hunziker et al., 2012), fluid balance (Lee et al., 2014), and adverse effects of medications (Danziger et al., 2013). We have also developed data access tools for MIMIC-II (Scott et al., 2013) as well as a machine-learning-based algorithm that can predict impending hypotension (Lee and Mark, 2010).

More recently, our focus has been on developing personalized patient outcome prediction algorithms that utilize patient similarity metrics (Lee et al., 2015; Lee, 2016).