By Dr. Courtney DiNardo
A few weeks ago, after I started one of my leukemia patients at MD Anderson Cancer Center on a standard course of chemotherapy, my patient developed a potentially life-threatening complication that sometimes occurs during leukemia treatment. It’s called tumor lysis syndrome. If not treated proactively, it can cause kidney failure, a heart attack and even death. A computing system based on IBM’s Watson technology that we’re currently piloting alerted me to the situation. I took action immediately. He’s okay now.
At an advanced cancer treatment center like MD Anderson, we likely would have spotted my patient’s problem early enough to respond in time without the help of a computer. However, in a community hospital, physicians who don’t see as many leukemia patients or have our expertise might not have noticed in time. The technology will definitely save lives.
This is a dramatic example of how cognitive computing technology has the potential to help improve medical care—benefitting not just patients and doctors but the entire healthcare system. I believe that the technology will help me accomplish my long-term career goal as a physician: to know enough about a particular patient’s cancer to provide exactly the right treatment for that individual. It’s the ultimate in personalized medicine.
My group helped train Watson to play its advisory role at MD Anderson. We are now in the testing and evaluation phase of this technology in the clinic with our patients. Watson will never take the place of a physician, but it can provide us with a treasure trove of information and expert guidance.
This approach offers several benefits. The system gathers information and presents it to the physician in summary form—but then allows you to drill down deep to see the evidence upon which it bases its recommendations. It provides an expert second opinion, if you will.
It’s a time saver. Last week, when one of my colleagues was out of town, I filled in and met with one of his patients—one with a complicated condition that needed a management decision. Under normal circumstances, it may have taken me all afternoon to prepare for the meeting with enough insight to provide the most appropriate treatment decisions. With Watson, I am able to get a patients history, characteristics, and treatment recommendations based on my patients unique characteristics in seconds.
The system is a health monitor. As I described at the beginning of this blog post, Watson can keep track of streams of information about a patient’s condition and alert physicians when something seems to be amiss.
The Watson system can also match individual patients who have not responded to conventional treatment in clinical trials that might be appropriate for them.
As we physicians interact with Watson, it learns and gets better at its job. There’s an annotation feature that allows us to type feedback to the system when it doesn’t draw on all of the evidence we need when we’re considering a diagnosis or treatment issue.
In the future, I expect this technology to help us discover new and better treatments for specific genetic abnormalities or associations of genetic changes. It will be able to help us evaluate rare or complex patient conditions and identify drugs that have already been approved for other uses that might also help in the situations we’re dealing with.
After spending much of my life training to be an oncologist, I’m now operating at the cutting edge of leukemia research and treatment. I’m grateful to have a new generation of genetics and computing tools at my fingertips when I take on the challenges that come at us every day at MD Anderson. I look forward to the day when we can say that we have conquered cancer.
To learn more about the new era of computing, read a free chapter of the new book, Smart Machines: IBM’s Watson and the Era of Cognitive Systems.