IBM’s Watson: Cancer Killer?

IBM has begun integrating its Watson supercomputing platform into a variety of industries, most notably healthcare: last week, Memorial Sloan-Kettering Cancer Center in New York City announced that, for the past year, it has partnered with IBM and WellPoint to train Watson in processing and interpreting oncology data.

IBM has long considered how to best use Watson, which boasts massive datasets and the ability to process natural-language queries—traits that could make it potentially invaluable for researchers, doctors, and other workers. Healthcare providers could upload their electronic records to Watson; combined with medical sources (such as journals and data from clinical trails) and human-machine training, that could make the platform into a cancer-hunting tool.

Indeed, clinicians and other human trainers spent nearly 15,000 hours over the course of the past year “teaching” Watson how to interpret clinical information. They also fed the supercomputer more than 600,000 pieces of medical evidence, along with two million pages of text from 42 medical journals and clinical trials. In addition, Watson can access 1.5 million patient records extending back decades, including patient outcomes, in order to produce insights into the best possible course of treatment.

Clinicians and analysts at Memorial Sloan-Kettering fed Watson with some 1,500 lung-cancer cases as a start to the platform’s ability to extract and interpret everything from physician notes and lab results. WellPoint provided, via its utilization-management pilot, more than 25,000 test case scenarios and 1,500 real-life cases, which apparently helped it analyze queries in the context of both complex medical data and natural language.

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

The Maine Center for Cancer Medicine and WESTMED Medical Group are apparently next in line to test the more oncology-minded Watson. IBM and its two partners also plan on issuing a set of Watson-based products that will allow researchers to dig into clinical datasets in order to create evidence-based decision support systems.

The products include Interactive Care Insights for Oncology, along with the WellPoint Interactive Care Guide and Interactive Care Reviewer. Interactive Care Insights for Oncology will be available via the cloud and give researchers and doctors a detailed record of data used to arrive at various treatment options. The other two products will use Watson to speed the review process between a physician and the patient’s health plan, potentially reducing waste, as well as analyze treatment requests and match them to medical policies and clinical guidelines.

The healthcare field is ripe for some supercomputing help. Think about all the complex factors that go into proper diagnosis, much less treatment—or all the patient data held by hospitals and other care facilities, which must be analyzed for useful insight. For some hospitals, Watson could indeed be the tool for cutting through at least some of this complexity.

 

Image: Franck Boston/Shutterstock.com

Post a Comment

Your email address will not be published.