Our genomics dataset consists of samples drawn from patients enrolled in the PREVAIL study, a phase II/III randomized controlled trial examining the role of lactoferrin in the prevention of nosocomial infection in the ICU. Patients have between 1 and 5 samples drawn for gene expression analysis (baseline, day 3, day 7, day 14, day 21, and day 28). Expression profiles are generated from whole blood, using the Affymetrix PrimeView array. Our current dataset includes approximately 200 gene expression profiles, taken from 70 ICU patients, and 20 healthy controls.

Our waveform dataset contains high frequency physiologic waveform recordings from all 33 beds in our ICU. High frequency waveform recordings (up to 500 Hz) are captured from the bedside monitors in the ICU of Kingston General Hospital. Waveforms include electrocardiography, pulse oxymetry, and invasive blood pressures. Our waveform data repository is housed at the Centre for Advanced computing, and currently includes more than 3,500 recordings totalling more than 15 TB of data. 

Our wearables dataset includes heart rate, activity, and sleep data from 50 patients in the ICU who wore the Fitbit Charge HR fitness tracker for a 24 hour period. A preliminary analysis of wearable-derived sleep data can be found here.

Screenshot of the lab's Fitbit data capture interface for clinical research

Screenshot of the lab's Fitbit data capture interface for clinical research


We use gene expression and clinical data from the Inflammation and Host Response to Injury ("Glue Grant") study.

We use the MIMIC database for studies on large scale clinical data. This popular repository includes clinical data from some 30,000 patients cared for in the ICUs at Beth Israel Deaconess Medical Center in Boston, and is maintained by the Lab for Computational Physiology at MIT.