The artificial intelligence known as IBM Watson can scan enormous troves of written information thanks to its natural language processing skills, and its machine learning programming means it quickly gets better at using that information to complete a given task. Most famously, it quickly got better at answering Jeopardy questions, and tromped the human competition in a 2011 exhibition match. More recently, IBM has been promoting the AI as the killer app for health care, where so much information is contained in written medical records and medical journal articles. Several hospitals and research institutions are testing Watson’s abilities to suggest personalized treatment plans for cancer patients.
At the Mayo Clinic, Watson will start by analyzing the medical records of patients with breast, colorectal, and lung cancer. (If all goes well, other patients will gradually be included in the project.) Watson will also be continuously scanning databases that list clinical trials, such as ClinicalTrials.gov, and will suggest appropriate matches for patients. There will be a lot to look through: The Mayo Clinic has about 8,000 clinical trials going on right now, in addition to the 170,000 that are ongoing worldwide. Mayo doctors will start consulting Watson in early 2015.
IBM vice-president of healthcare Sean Hogan says this system will provide new treatment options and new hope for patients, and will also speed the pace of medical research. And once Watson gets to work, it should get better and better at its job. “It’s designed to learn and improve,” he told IEEE Spectrum. “As it gets the iterative feedback, as it interacts with the experts, it gets better.”
Eliza Strickland is a senior editor at IEEE Spectrum, where she covers AI, biomedical engineering, and other topics. She holds a master’s degree in journalism from Columbia University.