We can all picture a hospital neonatal environment where a plethora of medical monitors connected to babies are used to alert hospital staff to potential health problems before patients develop clinical signs of infection or other issues. There are breakthroughs on the horizon for how this will be done. Today the instrumentation generates huge amounts of information — up to 1,000 readings per second — which is summarized into one reading every 30 to 60 minutes. The information is stored for up to 72 hours and is then discarded. If the stream of data could be captured, stored and analyzed in real time there would be a huge opportunity to improve the quality for special care babies.
The Hospital for Sick Children in Ontario, Canada has developed such a vision and is acting on it. Dr. Carolyn McGregor, Canada research chair in health informatics at the University of Ontario Institute of Technology visited researchers at the IBM T.J. Watson Research Center who are working on a new stream-computing platform to support healthcare analytics. A three-way collaboration was established, with each group bringing a unique perspective — the hospital focus on patient care, the university’s ideas for using the data stream, and IBM providing the advanced analysis software and information technology expertise needed to turn the vision into reality. The result of the collaboration was Project Artemis which pairs IBM scientists with clinicians and`researchers to explore how emerging technologies can solve real-world business problems, in this case developing a highly flexible platform that aims to help physicians make better, faster decisions regarding patient care for a wide range of conditions. At the Children’s hospital the focus is real-time detection of the onset of nosocomial infection (often called hospital-acquired infection).
Regulatory, ethical, privacy, and safety issues were addressed and then two infant beds were instrumented and connected to the system for data collection. The team then created an algorithm that describes the streaming data. By establishing the impact of moving a baby or changing its diaper those things can be filtered out to help spot the telltale signs of nosocomial infection.
Dr. Andrew James, staff neonatologist, at the Hospital for Sick Children is optimistic that as they learn more they will be able to account for variations in individual patients and eventually be able to integrate data inputs such as lab results or observational notes. In the future any condition that can be detected through subtle changes in the underlying data streams can be the target of the system’s early-warning capabilities. It is likely sensors attached to or even implanted in the body will allow monitoring of important conditions from home or anywhere.