IBM Watson for Oncology
San Antonio, TX. – Originally built as a complement for a physician and not as a replacement, IBM Watson for Oncology launched a study to determine how closely the artificial intelligence computer and a panel of 12-15 multidisciplinary tumor board members could agree on treatment approaches and how quickly they could reach a decision for patients.
We wanted to know more about how it would impact oncologists’ day-to-day practice and to assess how Watson’s recommendations compared to the decisions of our team of experts. – S.P. Somashekhar, MD, an oncologist and the study’s author and chairman of the Manipal Comprehensive Cancer Center, Manipal Hospitals, in Bengaluru, India.
IBM Watson for Oncology’s recommendations came in three categories: recommended standard treatment; treatment for consideration and not recommended.
Out of 638 cases of breast cancer patients, 90% of IBM Watson for Oncology’s recommendations were in line with the board.
While the recommendations were an important part of the study, they also measured the speed of the decision between Watson and the board. The study showed an average of about 12 minutes for the board to reach a treatment decision while Watson took a median 40 seconds to come to a recommendation.
Read more about the study and detailed results here.
Why This Matters:
Artificial Intelligence continues to emerge within the healthcare space and adding value in places healthcare was not expecting. Analyzing data at the speed of ‘light’ will only enhance a physician’s offering. With the mass amounts of data available, complements like IBM Watson for Oncology will be a welcomed addition to any ‘staff’. It will be interesting to watch how these systems are brought to market in an operational and manageable way and help further cancer care.