Designing Reliable Cohorts of Cardiac Patients across MIMIC and eICU.

The design of the patient cohort is an essential and fundamental part of any clinical patient study. Knowledge of the Electronic Health Records, underlying Database Management System, and the relevant clinical workflows are central to an effective cohort design. However, with technical, semantic, and organizational interoperability limitations, the database queries associated with a patient cohort may need to be reconfigured in every participating site. i2b2 and SHRINE advance the notion of patient cohorts as first class objects to be shared, aggregated, and recruited for research purposes across clinical sites. This paper reports on initial efforts to assess the integration of Medical Information Mart for Intensive Care (MIMIC) and Philips eICU, two large-scale anonymized intensive care unit (ICU) databases, using standard terminologies, i.e. LOINC, ICD9-CM and SNOMED-CT. Focus of this work is lab and microbiology observations and key demographics for patients with a primary cardiovascular ICD9-CM diagnosis. Results and discussion reflecting on reference core terminology standards, offer insights on efforts to combine detailed intensive care data from multiple ICUs worldwide.

The PhysioNet/Computing in Cardiology Challenge 2015: Reducing False Arrhythmia Alarms in the ICU.

High false alarm rates in the ICU decrease quality of care by slowing staff response times while increasing patient delirium through noise pollution. The 2015 Physio-Net/Computing in Cardiology Challenge provides a set of 1,250 multi-parameter ICU data segments associated with critical arrhythmia alarms, and challenges the general research community to address the issue of false alarm suppression using all available signals. Each data segment was 5 minutes long (for real time analysis), ending at the time of the alarm. For retrospective analysis, we provided a further 30 seconds of data after the alarm was triggered. A collection of 750 data segments was made available for training and a set of 500 was held back for testing. Each alarm was reviewed by expert annotators, at least two of whom agreed that the alarm was either true or false. Challenge participants were invited to submit a complete, working algorithm to distinguish true from false alarms, and received a score based on their program's performance on the hidden test set. This score was based on the percentage of alarms correct, but with a penalty that weights the suppression of true alarms five times more heavily than acceptance of false alarms. We provided three example entries based on well-known, open source signal processing algorithms, to serve as a basis for comparison and as a starting point for participants to develop their own code. A total of 38 teams submitted a total of 215 entries in this year's Challenge.