eICU Research Institute
Analytics to drive AI development and evidence-based best practices in the Intensive Care Unit
Critical care and high acuity: predictable and proactive
In the ICU, patients can become emergent quickly, and their care needs are constant and everchanging. This demanding care setting requires constant patient evaluation and care from the clinical staff. The Philps eICU program uses evidence-based decision-support tools, comprehensive analytics and specific predictive algorithms to provide data the virtual care team needs to help improve patient outcomes.
Features
Committed to sharing knowledge and the advancement of medicine
Philips eICU programs generate an incredible amount of data on ICU patient stays every year. Philips eICU Research Institute (eRI), a non-profit institute established by Philips and governed by customers, is a platform built from a repository of de-identified data, collected since 2008, that is used to advance knowledge of critical and acute care. For Philips customers that participate, eRI provides a unique platform to create one of the most comprehensive databases of ICU care in the world. In addition, eRI directly benefits customer telehealth programs as research is frequently translated into new and advanced tools and programs.
A comprehensive dataset helping researchers solve real world challenges in critical care
Philips established the eRI platform as a key enabler for critical research in the intensive care field. The database is a repository of de-identified data collected in collaboration with our customers. This integrated dataset contains billions of high quality representative clinical data points that extend over more than 15 years including the COVID-19 pandemic period. The secure database is instrumental in product development and includes detailed clinical information such as vital signs, pharmacy and medication orders, lab results, diagnoses, and novel severity of illness scores. The dataset gives comprehensive insights on patient admissions, treatments, co-morbidities, readmissions, and clinical outcomes.
Unlocking the value of eRI: advancing critical care through collaborative research and innovation
The impact of eRI extends from product development to critical research and beyond, enabling academia and member organizations to collaborate, innovate and advance critical care together. This database has played a pivotal role in developing solutions for multiple clinical challenges, including epidemiology/large data analytics, AI and predictive analytics and learning data patterns and clinical strategies.
Related offerings
Frequently asked questions
- Can researchers/clinicians access a sample of this data for research and development?
-
Philips is in collaboration with the Institute for Medical Engineering and Science (IMES) at the Massachusetts Institute of Technology (MIT) to allow health care researchers access to a large scale critical care dataset: eICU Collaborative Research Database eICU-CRD of 200K patients to help advance machine learning and artificial intelligence (AI) in healthcare. This is a subset of the large scale eICU Research Institute dataset mentioned above. The original eICU-CRD dataset was first released in 2016. More than 3,000 users have used the original database with citations in over 660 published academic research papers. A new dataset is released in 2023 with data from 2020-2021 including data acquired during the COVID-19 pandemic.
Documentation
- Published research
-
Epidemiology/Large data analytics:
- Association of systolic, diastolic, mean, and pulse pressure with morbidity and mortality in septic ICU patients: a nationwide observational study, 2023 (Annals of Intensive Care)
- Impact of COVID-19 pandemic on severity of illness and resources required during intensive care in the greater New York City area, 2020 (medRxiv)
- Five-Year Trends of Critical Care Practice and Outcomes, 2017 (Chest)
AI and predictive analytics:
- Improving ICU Risk Predictive Models Through Automation Designed for Resiliency Against Documentation Bias, 2022 (Critical Care Medicine)
- Early prediction of hemodynamic interventions in the intensive care unit using machine learning, 2021 (Critical Care)
- Predicting infection type upon clinical suspicion of hospital-acquired infection, 2020 (Critical Care Medicine)
- Estimating excess ICU time: potential financial impact and variation between ICUs, 2020 (Critical Care Medicine)
Learning data patterns and clinical strategies:
- Predicting the risk of mortality and readmission in the ICU, 2012 (Plos one)
- White papers
-
