IBM's Watson for Oncology in First Community Hospital

Nick Mulcahy

February 02, 2017

IBM's Watson for Oncology will be soon working with community oncologists to aid in the care of their cancer patients, as the data platform is being installed in Jupiter Medical Center, Florida.

The arrangement, which starts in March, is notable because it is the first community hospital to use the cognitive computing platform for oncology.

The Watson platform is probably still best known for its appearance on the American television quiz show Jeopardy!, where it beat human contestants.

The platform, which has since been "trained" at Memorial Sloan Kettering Cancer Center in New York City, has been used in oncology centers in the United States, including the Maine Center for Cancer Medicine and Blood Disorders, in Scarborough, since at least 2013, as reported by Medscape Medical News.

More recently, Watson for Oncology made news when it provided treatment recommendations for 638 breast cancer cases that were 90% concordant with the treating oncologists' opinions in a double-blinded validation study at a top cancer center in India, as reported by Medscape Medical News.

However, that impressive result came in the latter stages of the 3-year study after 175 "discordant cases" were reviewed a second time, adjusted by the oncologists, and were then once more passed through Watson.

In the study's first pass, Watson and the multidisciplinary tumor board at the Manipal Comprehensive Cancer Center in Bengaluru only agreed 73% of the time. The level of concordance also varied by breast cancer type, with higher agreement in more common types and lower unity in less common types, such as triple-negative disease.

In a press statement, Abraham Schwarzberg, MD, chief of oncology at Jupiter Medical Center, said, "We were impressed by Watson's analytical ability to help provide relevant treatment options for patients to allow physicians to personalize patient care in an unparalleled way."

Watson for Oncology uses more than 300 medical journals, more than 200 textbooks, and nearly 15 million pages of text to identify and rank evidence-based treatment options, including specific drugs and related administration instructions, according to the statement. Watson also links to peer-reviewed studies and clinical guidelines. Its "machine-learning" capability allows for continuous updating as new data accumulate.

Currently, Watson for Oncology can assist clinicians with treatment plans for breast, lung, colorectal, cervical, ovarian, and gastric cancers. IBM plans to train Watson on at least 10 additional cancer types in 2017.

Follow Medscape senior journalist Nick Mulcahy on Twitter: @MulcahyNick

For more from Medscape Oncology, follow us on Twitter: @MedscapeOnc


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.