Advanced stream computing software developed by IBM Corp. will be applied to a health care research project by the University of Ontario Institute of Technology (UOIT) that aims to help doctors detect subtle changes in critically ill premature babies.
The software should help medical practitioners make better decisions by observing and collecting physiological data streams such as respiration, heart rate, and body temperature, along with environmental data gathered from sensors, and combine it with data reaped from traditional paper-based methods.
The technology essentially allows researchers to create algorithms that can process streaming data and combine those algorithms in a multitude of ways in order to process incoming data in real time.
The goal of the research isn’t to replace the current manual paper-based approach employed by many medical practitioners. Rather, the technology will supplement those methods, said Carolyn McGregor, associate professor with UOIT and Canada Research Chair in Health Informatics. “That extra layer really represents a change in the fundamental way that critical health care is [administered],” she said.
Data collected through advanced stream computing will also be applied to findings from McGregor’s intensive care research. Life-threatening conditions, like infection, can be detected up to 24 hours in advance in premature infants by observing subtle physiological changes, said McGregor.
The neonatal intensive care unit at Toronto’s Hospital for Sick Children will be the first to deploy the technology, followed by the IWK Health Centre in Halifax, N.S., and another unit in Australia.
But before then, the software will be subject to a year-long test period at UOIT’s Health Informatics Laboratory using data already collected from neonatal intensive care units.
The Hospital for Sick Children was chosen for the initial rollout, McGregor said, because the neonatal intensive care unit often encounters complex medical cases, and experiences a high prevalence for the particular conditions in infants that the research team is interested in observing.
When fully developed, the software will be capable of processing the 512 readings per second generated by medical devices.
IBM has granted the researchers access to the prototype software as part of the company’s First of a Kind program, an initiative that connects research teams with customers “so that researchers are motivated by problems faced by IBM’s customers,” said Maria Ebling, senior manager and research staff member with the IBM. T. J. Watson Research Centre in Hawthorne, N.Y.
The relationship between IBM and UOIT, said Ebling, stemmed from a meeting at an academic conference last fall. UOIT learned of the stream computing software and “were excited by the technology and that’s when we started pursuing building the First of a Kind proposal.”
“It’s a collaboration where we help our customers work on problems that matter to them and at the same time on problems that matter to IBM,” she said.
If successful, the stream computing technology will be expanded to other neonatal intensive care units across the country through an affiliation UOIT has with The Canadian Neonatal Network, which comprises 29 units. The technology will be expanded across Australia as well, and eventually to Sri Lanka and China.
Although stream computing has obvious benefits for the health care market, Ebling sees applicability in areas like finance as well, where significant amounts of data are generated and analyzed. In fact, IBM has several ongoing pilot projects in sectors outside healthcare, including with TD Bank Financial Group.
While the advantages of the technology to neonatal intensive care are evident, McGregor said the lack of such systems among medical practitioners today is probably because “there isn’t the awareness within health and medicine of what technologies are out there and what the potentials.”
