Can AI improve cancer care in remote areas? 3 questions with Dr. Andrew Norden of IBM Watson Health

Jessica Kim Cohen - Print  | 

IBM has rolled out its oncology-focused clinical decision support platform in more than 55 countries in Asia, North America, South America, Europe and Australia since striking an agreement to develop the technology with New York City-based Memorial Sloan Kettering Cancer Center in 2012.

But the cognitive computing platform, Watson for Oncology, isn't just for academic medical centers. In a qualitative study out of the Instituto Nacional de Ciencias Médicas y Nutrición and Instituto Nacional de Cancerología in Mexico City, researchers determined the platform would be "particularly beneficial" for remote clinics that lack subspecialist expertise.

The study, presented at the American Society of Clinical Oncology's annual meeting in June, asked a panel of oncology experts in Mexico to provide feedback on Watson for Oncology. The panel considered the platform's recommendations for 100 cases related to lung, breast, gastric, colon and rectal cancer, and concluded it would be "very useful" for improving consistency, efficiency and overall quality of cancer care in Mexico.

"There's now a growing body of knowledge that shows these tools today are helping providers and patients all over the world," says Andrew Norden, MD, deputy chief health officer for oncology and genomics at IBM Watson Health. "I think we're at an early time in artificial intelligence tools for cancer care, but my prediction is, in five or 10 years, these tools will be ubiquitous."

Dr. Norden, a practicing neuro-oncologist, spoke with Becker's Hospital Review about how artificial intelligence may improve cancer care across the globe.

Editor's note: Responses have been lightly edited for length and clarity.

Question: I've been following IBM Watson Health, and it looks like oncology has been one of its primary focuses. Why oncology?

Dr. Andrew Norden: Oncology is such an important, challenging and rapidly changing area of healthcare. There is good evidence that too much of the time healthcare decisions — and oncology decisions, in particular — aren't based on evidence. A substantial amount of the time we think it's because even though evidence exists, physicians aren't aware of it. So many things are changing, almost month by month: new drugs, new categories of drugs, new cellular therapies, new paradigms in which we treat cancer based on molecular data. It's enormous and challenging to shift through, and even the best-trained physicians just can't keep up.

We think one way to ensure we progress is by giving physicians the tools they need to make the best possible decisions at the point-of-care. We're suffering from information overload, and I think it's abundantly clear artificial intelligence or cognitive computing systems like Watson have the capacity to bring the right information to the physician to help them at the time they're making a decision. It's also worth mentioning cancer is a worldwide problem. Although progress has been made in recent decades, much of that progress has been in the United States. We're trying to build tools that help improve oncology care anywhere in the world.

Q: IBM presented a few international studies about Watson at ASCO. Could you speak to the significance of the qualitative research paper out of Mexico?

AN: That point from the Mexican oncologists is really in agreement with what we're hearing from physicians around the world. The amount of information available about cancer care and optimal therapies today is exploding, and it's really challenging for physicians to keep up with. If you work in an urban academic center, like the kind where I came from, you tend to focus on a very narrow therapeutic area. I was a brain tumor specialist, for example. Among my colleagues, some of us focused on only one specific type of brain tumor. I can tell you, even working in that type of narrow area, it is challenging to keep up with the amount of information.

Now, imagine you are one of the general medical oncologists working in Mexico. You are likely seeing patients with breast cancer, lung cancer, prostate cancer, colon cancer, cervical cancer — just to name some of the common ones — and, of course, a patient with a rare cancer subtype can walk in at any time. The likelihood you, in that general medical oncology clinic, are able to keep up with the literature in each of those fields, is extremely small. Subspecialist physicians might feel overwhelmed by the amount of information they need to master, but generalists feel exponentially more overwhelmed. I think the potential impact of decision support tools like Watson are even greater in that setting.

Q: If there was one thing you could tell hospital leaders about the role of artificial intelligence in the healthcare setting, what would it be?

AN: One point I'd like to share relates to one of the biggest misconceptions: that artificial intelligence technologies are likely to replace physicians. We see technology as a real partner in the care of cancer patients, and we see how physicians, enabled by technology solutions, are able to make optimal decisions based on data and available evidence. There are some key things artificial intelligence systems excel at, and may do better at than physicians, like finding meaningful patterns and sorting a nearly limitless amount of data without having to sleep. However, humans are critical to the care of cancer patients.

I don't foresee any capacity for computers to sit face-to-face with a patient and talk about their individual hopes, dreams and fears related to cancer, or to help a person talk through the pros and cons of different treatment options, delving deep within the individual to see what's most important to them as their cancer treatment unfolds. From our perspective, this is a partnership between human and machine, and I think you're seeing a growing body of evidence that this really can make a meaningful impact on cancer patients and their families in every corner of the world.

More articles on health IT:
UChicago Medicine and Google — a data-driven duo to watch
6 thoughts on population health with Allscripts Analytics CMO Dr. Fatima Paruk
Paper records are here to stay: 4 questions with Illinois Pain Institute's Dr. John Prunskis

© Copyright ASC COMMUNICATIONS 2019. Interested in LINKING to or REPRINTING this content? View our policies by clicking here.

To receive the latest hospital and health system business and legal news and analysis from Becker's Hospital Review, sign-up for the free Becker's Hospital Review E-weekly by clicking here.