Octo Consulting, a technology solutions and management services company for both the intelligence and healthcare sectors, recently published an infographic exploring the intersection among the Semantic Web, Linked Data and Health IT as it relates to accessing and interacting with data from an array of sources in the healthcare chain. “Our point of view is that in healthcare there are multiple data sources and so much data – especially when it comes to clinical trials, pharmaceuticals research and scientific data,” says CTO Ashok Nare. “It’s very possible that each of those data elements is represented in a different format, so how to take them all and connect them to ask questions you aren’t able to ask otherwise. That’s where semantic technologies are extremely useful.”
One health-care sector project in which Octo is putting semantic technologies to use these days is an effort it has underway with the U.S.’s medical research agency, the National Institutes of Health, whose mission includes providing grants to the scientific community to engage in research and experiments “to enhance health, lengthen life, and reduce illness and disability,” as its web site explains. Now, not only does the NIH want to understand what it’s funding and how those grants are progressing, but also “what opportunities it may be missing out on,” Nare explains.
That means continually assessing not only what’s in its portfolio but also what research gaps there are, which requires conducting analysis on more and more data sources and investigating more queries: That could mean more development and expense, without the help of semantic web technologies.
“As you expand the query set into what are the gaps in the portfolio, to do broader portfolio analysis, that requires a different perspective and forces us to bring in more data sets,” he says. Octo Consulting started the conversation with NIH about using semantic technologies to aid in the process awhile back, discussing the benefits of leaving the data where it lives and linking it together to answer questions and correlate with other queries rather than using data warehouse technologies that require explicitly loading different data sets for every new query.
Last year it advanced those discussions around linking data sources and building inferences and reasoning layers with a couple of prototype use cases. As the infographic below illustrates, this included laying out the differences in using a data warehouse versus semantic web technologies and linked data to look at different clinical trials around cancer drug discoveries by investigating five different data sets.
“Right now we are trying to leverage all the background we have with this customer in their understanding of semantic technologies to look at the market, and what different solutions are available for connecting these data sources,” Nare says. “We want the customer to be able to ask any different type of question and not need different toolsets to do that.”
Octo is working to develop additional prototypes and provide more information to NIH to show the value of semantic technology in accomplishing its ends. That includes looking at what is a growing range of solutions and advancements in the space that will help cement semantic technology’s role at the NIH. “We want to leverage the latest and greatest,” he says, including solutions available from the open source community.
Says Nare, “if we use semantic tech it’s not only saving money and investments in processing and answering questions, but also can be in building the solutions themselves. That’s where the wins are.”