Royal Philips announced an expansion of its initiative with the Institute for Medical Engineering and Science (IMES) at the Massachusetts Institute of Technology (MIT) to allow health care researchers access to a new critical care data set to help advance machine learning and artificial intelligence (AI) in healthcare. The updated eICU Collaborative Research Database (eICU-CRD) includes de-identified data of 200,000 critical care patients, including patients who were impacted by COVID-19. The broader and clinically dependable data set will support the development of solutions that improve patient care and clinical outcomes.
During the COVID-19 pandemic, eICU and critical care saw a dramatic increase of patients and unique challenges in the way that care was provided, prompting Philips and IMES to expand the original data set, first released in 2016. The new secure database includes de-identified and detailed clinical information such as vital signs, pharmacy and medication orders, laboratory results, diagnoses, and novel severity of illness scores. The dataset gives comprehensive insights on patient treatments, co-morbidities, readmissions, and clinical outcomes.
Researchers at Philips and the Laboratory of Computational Physiology within IMES will grant researchers around the world access to the data to help develop advanced algorithms and provide new insights on critical care. The Laboratory of Computational Physiology will continue to serve as the academic research hub for the initiative and will provide and maintain access, as well as help educate researchers on the database and offer a platform for collaboration. The database is available for medical research, to those who are credentialed, who take human subjects training, and who agree to a data use agreement.
Jesse D. Raffa
research scientist in the Lab for Computational Physiology at IMES
“The database, which includes patient information from 2020 and 2021, now contains significant overlap with the Covid-19 pandemic, yielding valuable patient data for research,” said Leo Anthony Celi, principal research scientist and clinical research director at the Laboratory of Computational Physiology at IMES. “This updated database is a vital resource for education, including in many courses at institutions like Harvard, MIT and Stanford; and training, as well as low-resource institutions,” said Jesse D. Raffa, research scientist in the Lab for Computational Physiology at IMES.
The eICU-CRD is the only dataset containing detailed critical care data from over 200 hospitals across the U.S., representing many ‘real-world’ challenges for successful deployment of algorithms and models, which are often not readily apparent in single-center datasets. Unlike other organizations that do not share data or only share single source data sets, Philips shares its data with credentialled researchers to help advance AI for improving outcomes in human health. More than 3,000 users have used the original database with citations in over 660 published academic research papers, including in Nature, The New England Journal of Medicine and the Journal of the American Medical Association.
“This initiative demonstrates our commitment to advancing machine learning and AI efforts, by making eICU data available for global research initiatives,” said Shiv Gopalkrishnan, General Manager of EMR & Care Management at Philips. “This is how we can enhance patient care and improve clinical outcomes: liberating and connecting data across systems and applications with integrated devices, systems and informatics, which can inform research with patient insights that can help clinicians make the right decision at the right time for their patients.”