There’s a new term in town: Participatory mHealth. As defined by Deborah Estrin, Professor of Computer Science at CornellNYC Tech and co-founder of Open mHealth, it “is taking what was previously unmeasured and uncaptured behavior – [things] that were previously ephemeral — and turn that into data.” Such information can be as valuable to treating conditions as is data collected by the instruments and diagnostics tools in formal health care settings, and perhaps can be a more reliable indicator of how a person’s health really is faring than their response to a doctor’s question at exam time. Those answers, after all, can be influenced by so many things – how they’re feeling that morning, for instance, vs. how they’ve generally felt since the last time they spoke to their healthcare provider.
With so many people today in possession of a smart phone with built-in GPS capabilities, there’s a new opportunity to capture so much data about what individuals do – especially those with chronic conditions – as well as the distance parameters related to where they’re doing it, the times they set out for an activity, how they’re feeling at various times during the day, and a whole lot more. “Chronic disease in some ways is the killer app for this kind of mobile health technology,” Estrin noted, since most of the care for dealing with chronic conditions occurs outside the clinical setting.
The capture is the easy part, says Estrin – much of it can be automated or be entered via a simple click, for instance. “The heavy lifting is in the analysis, in fusing and pulling out what is interesting from those data streams,” Estrin told an audience at the Semantic Technology & Business Conference in New York City on Wednesday.
The value is in the patterns, the trends, the anomalies that can be useful to individuals in their self-care and to their clinicians, as well. Research evidence is another potential workflow that can benefit.
“Health and health management is to some extent an optimization problem,” she said. “If you don’t have the right frequency and resolution of data in the right time, it’s hard to achieve [that optimization].”
How can the semantic community contribute to the open – and Big – Data picture that open mHealth is drawing? “It is finding the meaning in the data,” she explained. “It’s a combination of getting the syntax and the plumbing and the semantics right.” Open mHealth is about trying to create more modular, shareable and sense-making components to pull out meaning from data. For that, you’ve got to do things like get activity classification right, and more—“stuff your industry has grown up knowing about,” she said. It’s a hard ontology problem of interpreting, for example, what a glucose measure actually means.
Open mHealth envisions connecting innovators, ideas and techniques to create new tools, as its web site explains, integrating and complementing both proprietary and open-source work from the public or private sector. It’s a vision of small modules and simple APIs, where open source is used for the commodity pieces but innovations built atop that can be closed-source.
“We try to push the semantic and standardization and actual meaning as far out to the edges as possible, to the last possible moment so the inner plumbing has as little baked-in semantics as possible,” Estrin added at the event. “It’s a light- handed approach to semantics to foster a broad ecosystem of software components.”
- Session Spotlight: Incubating and Investing in Semantic Startups
- Status Update on US Open Data Collaboration
- Big Data Is Big Focus At SemTechBiz
- Big Data Skills Worth Big Bucks