IBM's Watson to Help Doctors Diagnose, Treat Cancer

 
 
By Brian T. Horowitz  |  Posted 2011-12-21
 
 
 

The next assignment for IBM's Watson supercomputer is to evaluate treatments for cancer. Health insurer WellPoint, which is looking into real-life applications for Watson, has announced that a leading cancer institution in Los Angeles, Cedars-Sinai's Samuel Oschin Comprehensive Cancer Institute, will provide expertise to the company on developing applications based on IBM's Watson data-analytics machine.

Cedars-Sinai is a medical facility focused on diagnosing and treating many types of cancers. WellPoint offers health plans such as Blue Cross and Unicare.

"As we design the WellPoint health care systems that leverage IBM Watson's capabilities, it is essential that we incorporate the highly specialized knowledge and real-life practice experiences of the nation's premier clinical experts," Dr. Harlan Levine, executive vice president of WellPoint's Comprehensive Health Solutions, said in a statement.

Watson allows doctors to understand the large amounts of clinical data and patient histories in medical libraries. The supercomputer gained fame on "Jeopardy" earlier this year and incorporates deep computing and data analytics as well as natural-language processing (NLP) capabilities from Nuance Communications.

WellPoint plans to expand the use of Watson to specialists such as cardiologists and pediatricians, Anthony Nguyen, WellPoint's senior vice president of care management, told eWEEK in September.

By working with WellPoint, IBM hopes to coordinate communication among health care providers, benefits administrators and patients.

WellPoint announced its collaboration with Cedars-Sinai Dec. 16.

A large amount of information and numerous sources of data make oncology an area that could potentially benefit from Watson, said Dr. M. William Audeh, medical director of the Samuel Oschin Comprehensive Cancer Center, which is part of the Oschin Institute at Cedars-Sinai.

"I think that's the kind of challenge that lends itself to a supercomputer like Watson, to try to integrate all of that information," said Audeh.

"What we're talking about is the genomic revolution; genomic information is now being applied more and more to cancer medicine," Audeh added. "New technologies for both diagnosis and treatment and targeted therapies are contributing to the large complexity, and that's why I think the time is right to apply a computer-based approach of the sort that Watson offers to integrate all of that information."

Although Cedars-Sinai won't use Watson itself, as part of the agreement, the institute will advise WellPoint on which clinical content to include in future applications built using the supercomputer, Audeh explained.

Selecting the information to feed into Watson is challenging and complicated.

"It's a very complicated process to, first of all, select the basic information that needs to go into Watson to make it well-informed about all the various kinds of cancer treatments and all the nuances of information that surround a patient with cancer," said Audeh. 

Using its data analytics and NLP capabilities, Watson would integrate data such as medical literature, patient histories, clinical trials, side effects and outcomes data to help doctors decide on courses of treatment.

"I think the hope is that Watson would be taught to integrate basic standard information from the medical literature, from clinical trials, from all the guidelines that have been developed by the medical societies into a database that can then look at an individual patient's particular characteristics," said Audeh.

Watson would also look at the characteristics of a patient's cancer and make recommendations on cost-effective treatment that would lead to the best outcome, he added.

"All of that [would] not replace a physician but provide a more comprehensive resource for all of the medically relevant information than any doctor can currently access on a computer, said Audeh.

With the many possible ways to go about cancer treatment, Watson technology could be helpful in evaluating these methods, said Audeh. 

WellPoint plans to launch an application based on Watson next year in clinical pilots.

In addition, IBM announced in May an expansion of its Analytics Solutions Center in Dallas to link health data to electronic health records (EHRs) on mobile devices using NLP and data analytics similar to that of Watson.

Other health care IT giants have also been active recently in developing technology to help find treatments for cancer. In November, Dell donated a cloud infrastructure to the Translational Genomics Research Institute (TGen) for storing data from a personalized medicine trial for pediatric cancer. 

In another development, General Electric pledged $100 million to cancer research over the next five years.

 


Rocket Fuel