Artificial Intelligence Gets Real

Focusing the power of IBM's Watson on helping patients

By Mayo Clinic Staff

Ask people what they know about Watson, IBM's cognitive computing system, and many might answer, "Didn't it cream those all-star contestants on Jeopardy a few years ago?"

The answer is yes, it most certainly dealt Jeopardy champs Ken Jennings and Brad Rutter an ego-bruising defeat in 2011.

So how did Watson do it? And how can this technology improve health care?

Watson is artificial intelligence in action. It mimics the way humans observe, interpret, evaluate and make decisions based on data. Because it can read everyday language, Watson can digest and understand vast amounts of information from books, articles, Web pages and other readily available sources in a single day. Importantly, the more Watson learns, the more its decision-making skills mature and improve.

This capability sparked a big idea at Mayo Clinic: What if Watson became part of the clinical team?

Through a pilot collaboration with IBM, Mayo researchers are finding out. This proof-of-concept study applies a customized version of Watson to match patients more quickly with appropriate clinical trials, beginning with cancer.

"In an area like cancer — where time is of the essence — the speed and accuracy that Watson offers will allow us to develop an individualized treatment plan more efficiently, so we can deliver exactly the care that the patient needs," says Steven R. Alberts, M.D., chair of Medical Oncology at the Mayo Clinic Cancer Center.

Clinical trials provide patients with access to new and emerging treatments, yet enrolling participants is one of the more difficult parts of research. The standard process is manual, with clinical coordinators sorting through patient records and conditions, trying to match patients with the requirements of a given study protocol.

The high volume of studies complicates matters further. At any given time, Mayo Clinic conducts more than 8,000 human studies in addition to the 170,000 that are ongoing worldwide.

Churning through and making sense of big data like this is what Watson does best. If the pilot is successful, it could unclog the enrollment bottleneck at Mayo and elsewhere. Despite Mayo Clinic's best efforts, just 5 percent of its patients take part in studies. Nationally, the rate is even lower, at 3 percent. Through initiatives such as Watson, Mayo hopes to double the number of patients it can help through clinical trials.

"Mayo Clinic is embracing cognitive computing because we realize this technology is transformational and necessary for the continued evolution of health care delivery," says Nicholas F. LaRusso, M.D., a Mayo Clinic gastroenterologist, the project lead for the Mayo-IBM Watson collaboration and the Charles H. Weinman Professor. "In this pilot alone, using Watson will allow Mayo's research and clinical teams to spend more time focusing on the needs of the patient rather than the laborious process of matching eligibility requirements to each patient's situation."

Your generosity fuels innovative approaches to patient care. Make a gift today.

March 01, 2016