Memorial Sloan-Kettering and IBM Watson

Memorial Sloan-Kettering Cancer Center and IBM today announced plans to develop advanced oncology decision support using IBM’s Watson supercomputer.  Researchers at IBM and the world-renowned hospital will input data from, among other sources, the cancer center’s Sunrise Electronic Health Record, to strengthen Watson’s capabilities as a medical resource. 

Martin Kohn, chief medical scientist at IBM, told the AP/Washington Post that, after the completion of the oncology training in 2013, Watson should be able to quickly recommend cancer diagnoses and provide treatment options.


Larry Norton, deputy chief for breast cancer programs at Sloan-Kettering, told the Post that the data will add “wisdom” to the computer’s grasp of scientific literature. He said, “Because of our size and experience, we have super-specialized physicians in every field of cancer. And all of what they actually do is capturable in the language of our electronic medical records.”

After inputting the oncology data, researchers will test Watson’s knowledge using complex cancer care scenarios. The supercomputer also will be assessed by an advisory board.

Watson gained notoriety last year when it defeated the top two contestants of the TV game show Jeopardy!, in one of history’s most famous human-to-machine matchups.  The artificial intelligence system uses advances in natural language processing and analytics to process information in a similar way that people think.

Tags: , , ,

About the author

Todd Stein is editor of It Takes A Community. A former journalist and freelance editor, Stein has worked as a business reporter for The Sacramento Bee and San Francisco Business Times, as producer of HIMSS Newsbreak, contributing editor to California Hospitals and senior writer for NurseWeek. His many articles on healthcare IT and other topics have appeared in dozens of national and regional publications including Healthcare Informatics, Healthcare IT News, Journal of AHIMA, Physicians Practice, The Los Times, San Francisco Magazine and VIA.



Dhruv Sakalley says:

03/23/2012 at 7:54 am

Great news! I was wondering if we could build in intelligence and decision support inside of our applications that cater to more personal charachteristics well defined in a patients’ profile, such as medical history and trends. A lot could be done in the area of personal medicine with the DNA sequencing technologies getting cheaper by the day. We already have the data, we would however need to work out the legal limitations of use of this data for statistical research.


Your email address will not be published. Required fields are marked *