IBM's Watson becomes a cancer treatment adviser

Supercomputer draws on troves of cancer research and patient records to guide doctors toward insurance-approved treatments

IBM's Watson is increasingly starting to resemble the fictional, hyper-intelligent supercomputers HAL from "2001: A Space Odyssey" and HOLMES from "The Moon Is a Harsh Mistress." Since defeating human players at "Jeopardy" in 2011, Watson has been groomed by IBM, Memorial Sloan-Kettering, and health-insurance giant WellPoint to provide personalized treatment recommendations for cancer patients in a matter of seconds.

"These breakthrough capabilities bring forward the first in a series of Watson-based technologies, which exemplifies the value of applying big data and analytics and cognitive computing to tackle the industries most pressing challenges," said Manoj Saxena, IBM General Manager for Watson Solutions.

From a strict technological standpoint, the new applications of Watson are impressive and groundbreaking: The system is capable of combing through 600,000 pieces of medical evidence, 2 million pages of text from 42 medical journals and clinical trials in the area of oncology research, and 1.5 million patient records to provide on-the-spot treatment recommendations to health care providers. According to IBM, more than 90 percent of the nurses who have worked with Watson follow the guidance the system gives to them.

However, the development leads down a slippery ethical slope, raising the question of just how much power and influence a soulless machine working on behalf of a for-profit health care companies should have in recommending, approving, or denying courses of treatment.

First, though, a brief primer on Watson: It's effectively a powerful analytical engine that combs through scores of data sources in real time, seeking out patterns and relationships among data sets to find solutions to questions. Watson's ability to analyze natural language, which was evident when it played "Jeopardy," is a key part of its success, enabling it to understand the crux of questions and to present clear, concise, and quick answers. Watson reputedly has a bright future in such fields as finance, city planning, and health care.

For more than a year, IBM has partnered separately with WellPoint and Memorial Sloan-Kettering to train Watson in the areas of oncology and utilization management. Clinicians and technology experts spent thousands of hours "teaching" Watson how to process, analyze, and interpret the meaning of complex clinical information using natural language processing in the name of improving health care quality and efficiency.

"It can take years for the latest developments in oncology to reach all practice settings. The combination of transformational technologies found in Watson with our cancer analytics and decision-making process has the potential to revolutionize the accessibility of information for the treatment of cancer in communities across the country and around the world," said Dr. Craig B.Thompson, president of Memorial Sloan-Kettering Cancer Center.

Memorial Sloan-Kettering and WellPoint are using Watson in different ways. Sloan-Kettering worked with IBM to develop its Interactive Care Insights for Oncology system. It's a cloud-based service, designed to provide medical professionals and researchers with helping in identifying individualized treatment options, based on a patient's medical information and the synthesis of various updated and vetted treatment guidelines and published research. It also provides users with a detailed record of the data and information used to reach the treatment options. The Maine Center for Cancer Medicine and WESTMED Medical Group are the first two early adopters of the capability.

WellPoint worked with IBM to develop two systems designed for utilization management: the WellPoint Interactive Care Guide and Interactive Care Reviewer. The systems are designed to analyze treatment requests and match them to WellPoint's medical policies and clinical guidelines "to present consistent, evidence-based responses for clinical staff to review, in the anticipation of providing faster, better-informed decisions about a patient's care."

During WellPoint's pilot program of these offerings, Watson absorbed more than 25,000 test case scenarios and 1,500 real-life cases, according to IBM, and gained the ability to understand and analyze queries in the context of complex medical data and human and natural language, including doctors' notes, patient records, medical annotations, and clinical feedback. Watson started processing common medical procedure requests by providers for WellPoint customers in December and was expanded to include five provider offices in the Midwest.

On paper, Watson appears to be a very powerful medical tool, providing doctors with unprecedented access to expert advice drawn from a vast trove of the most up-to-date medical research and case-study information. Health care organizations that embrace Watson will, of course, stress that it's just a tool. But will there come a point when doctors will give more weight to Watson's advice than to their own knowledge and instincts -- coldly calculated advice weighted by how much health insurance companies are willing to spend on treating a patient? "I'm sorry, doctor. I'm afraid I can't do that."

This story, "IBM's Watson becomes a cancer treatment adviser," was originally published at Get the first word on what the important tech news really means with the InfoWorld Tech Watch blog. For the latest developments in business technology news, follow on Twitter.

Copyright © 2013 IDG Communications, Inc.