• New Zealand (English)

Microsoft in Health

What does natural language processing mean for healthcare?

08 May 2014 | Tom Lawry, Global Product Strategy, Health Analytics

natural language processing (NLP)Most healthcare leaders I speak with consider their workforce to be knowledge workers. So I always like to ask them: What if you could give your knowledge workers access to knowledge at that “moment of need”? And by that, I don’t mean that health professionals have to ask an analyst and go through a long, laborious process to get answers to their data questions. I mean: What if they could easily do the research themselves to get the insight they need—when they need it—to help them do their job better?

With natural language processing (NLP) this isn’t just a futuristic notion anymore—analytics for the masses is the here and now.

NLP is a field of computer science, artificial intelligence, and linguistics concerned with the interactions between computers and human (natural) languages.

In the past, analyzing data required a specialist who knew how to speak in the computer’s language—using codes or formulas, for example—to illicit the desired insight. NLP is flipping that process on its head. It enables the computer to understand human language. It’s designed to simplify informatics so that anyone—not just a specialist—can access and analyze data. It’s basically a very intelligent system that takes advantage of things like machine learning, so the more queries you make, the smarter it gets at finding answers.

So what does that mean for healthcare?

NLP is truly liberating and democratizing data so that everyone across your health organization can have research on demand capabilities at their fingertips. For example, using the query and answer functionality in Power BI for Office 365, a health professional can simply type in a question, and NLP translates it into data analysis to return an answer almost instantaneously.

Not only that, Power BI brings back the answer in a format that’s easy to comprehend. Yes, there are some people who are PhDs in analytics and are wired genetically to really love tables with rows and rows of data—but most of us aren’t that way. So the Power BI team worked hard to make sure that not only can you find the data you need—as easily as typing a question into Bing search—you can then visualize it in a number of different ways. You can choose to see it in a pie chart or a bar graph, for example, to help you understand your research results. And the visualizations change dynamically as you modify the question so that you can have an interactive experience with your data.

The Power BI team recently indexed several massive databases, including 130 years of disease incidence data from the United States Centers for Disease Control, and data on every prescription written in Australia over a six-year period.

Imagine the effort it would take to sort through all that data using standard tools and queries? With Power BI Q&A, health professionals can simply type in a question such as: Which diseases have the highest incidence among Americans? Or: Which drugs in Australia are used the most? And they’re immediately presented with easy-to-understand data addressing their query.

These are just a couple examples. By enabling analytics for everyone, NLP offers tremendous potential for arming health professionals with on-demand insight to help them provide better care and work more efficiently.

Have a comment or opinion on this post or a question for the author? Please send us an email or let us know on Facebook or via Twitter.

Tom Lawry
Global Product Strategy, Health Analytics