Science and Serendipity Meet at the IBM Accelerated Discovery Lab
Illustration by akindo
Scientists, artists and entrepreneurs frequently don’t know when a bright idea might hit. It could be in the middle of the night, in the shower or over a breakfast of eggs and bacon. And the serendipity often involved is part of the beauty of discovery.
It also helps when groups work in concert with others. This speeds up the discovery process, with many minds working to solve a particular problem using toolsets they might not have previously considered.
And that’s essentially the idea behind IBM’s Accelerated Discovery Lab in San Jose, Calif. It brings clients and researchers together to tackle problems that seem intractable—and accomplish them more quickly than they could have on their own. As Laura Haas, IBM Fellow and director of the Accelerated Discovery Lab, explains, good ideas can come from anywhere.
Q: What’s the focus of the lab?
A: Our focus is really on helping people get actionable insight out of big data using analytics that will impact their businesses or their lives. We’re talking about big data analytics and this notion of accelerating discovery. We’re finding that cognitive or collaborative technologies are also very key to what we need to do here, so it’s more than just about a big data or analytic fact. It’s actually thinking about the future of how you can help somebody design their analytics so they get the value out of their data.
One of the things that we say in the lab is that the tools you already have can help you find what you already know. What our lab will help you do is discover what isn’t yet known, and that’s a next step in cognitive computing.
Q: How did the idea for the lab come about?
A: One of the motivations for the lab came from a meeting I had with medical researchers in Texas about potential research collaborations. I was there to talk with a number of companies and schools in the medical research environment about forming a center of healthcare analytics. That didn’t come to pass, but in the process I met a medical researcher from the Baylor College of Medicine, and I mentioned the IBM strategic IP insight platform (SIIP), which can be used to extract information on chemical compounds from patients and the medical literature.
The researcher himself was looking into potential cancer cures, and asked if the same sort of technology could apply to genomic information. He couldn’t keep up with the volume of new medical literature produced, even if the material was restricted to a single gene. But by knowing what was in the medical literature, he could focus his time and energy on the most promising leads.
In the Accelerated Discovery Lab, we used technology similar to Watson to read and digest that literature, form a model of it and then help him better focus his attention on the most likely candidates to control the gene. That was an aha moment of bringing together his medical knowledge, our IT knowledge and some predeveloped tools that might lead to some really important results.
Q: Is research in the lab different from research typically conducted within IBM?
A: In some ways it’s very different, and in some ways it’s totally the same. We do have a history of collaboration, which is often where we get inspirations that lead to new innovations and eventually new products and offerings for IBM. But there are two differences here. One is having an opportunity for a much closer engagement with other organizations than we may have had in the past. In typical engagements, the client doesn’t actually get to touch and see what we’re working on. They work through the researchers, but the researchers do the bulk of the work. Here, they can come in and really engage with us in collaborative innovation. Not all clients are going to want that, but it’s available now if they do. The other thing that I think makes this very different is the way that clients engage with IBM Research. They’re engaging with one narrow group that has a particular expertise or knows a particular technology very well. As a result, they can get a much closer engagement and interaction with the researchers.
comments powered by