How Collaboration Between IBM And Memorial Sloan-Kettering Taps The Wisdom Of Physicians

To me, the recently-announced collaboration between my institution, Memorial Sloan-Kettering Cancer Center, and IBM is of profound importance for reasons both obvious and more subtle. The obvious reasons concern what we may call knowledge. The subtle ones would be apparent only to those of us who care for patients on a daily basis and concern what we may call wisdom. Let me explain what I mean by the distinction.

First: What is the project? Our two organizations plan on applying IBM's Watson technology, which impressed all of us by beating two grand-champions on the Jeopardy! TV quiz show, to MSKCC's vast store of cancer case histories, utilizing the skills of expert computer scientists and highly experienced cancer doctors. The goal is to develop a tool to help physicians all over the world better care for their patients with cancer.

We live in a time of massive expansion of knowledge about cancer. Biological scientists all over the world, including at MSKCC, are producing information continuously, including information about the molecules that go awry to cause cancer in the first place. Furthermore, there is a steady stream of clinical research advances in the diagnosis and management of cancers of all types.

To be useful in patient management all of these bits of information need to be gathered and assimilated and made practical. For decades MSKCC has had specialized cancer physicians and scientists focused on the production and accumulation of such knowledge. In addition, we have used this information in the care of our patients and we freely share our knowledge as well. For example, colleagues around the world frequently call us to discuss difficult or unusual cases. By asking questions and exchanging information the treating physician and the MSKCC specialist arrive at a management plan that is best for that particular patient at that moment in time.

But this is not an efficient process. Most patients with cancer in the world are not treated by specialists in their type of cancer. The physicians treating them do have access to written guidelines and other sources of information, but that is not the same as having an experienced, specialized medical expert immediately available. Hence, we have long sought a means of bringing up-to-date knowledge to the bedside of every cancer patient.

But how can we do that when there is so much information, when it is constantly changing, when it needs to be interpreted, and when it is in the form of language with all of its subtleties and nuances?

Along comes IBM Watson.

Here we have the confluence of two highly developed areas of intellectual activity. At MSKCC we not only have extensive experience in the care of tens of thousand of cancer patients--how they have been treated and their outcomes as well--but we have this in written form. This is because of our use of a sophisticated electronic medical record system, meticulously constructed so as to capture all of the relevant clinical information in a way that absolutely protects patient privacy.

In Watson we have a computer system that can read and understand language, interact with human experts, and remember everything it has ever learned. And it can use this knowledge to arrive at answers to real-life questions.

As medical educators, we take young doctors and educate them to be expert cancer specialists in a few years, and we do this by teaching in English. Now we have a machine that can be taught in English and will never forget the knowledge we impart. And, by constantly learning, the machine will also produce new research questions that will help us improve the state of the art as we learn together.

But, one may ask, what about the human side of the equation? Medical decisions are not just about information: they are also about judgment. As experienced physicians we need to take everything into account in arriving at a best management plan--not just the individual patient's biology and the biology of their disease, but social, psychological, environmental, motivational and interpersonal factors.

It is here that the less obvious advantages to our project become paramount. These personal elements are captured in the language of our case histories, and Watson will learn them-while remaining ignorant of patient identity. Furthermore, Watson will not make decisions, but will interact with the on-site physician in asking the right questions to help the patient and the doctor arrive at the right decision for that particular individual. The human is never left out of the conversation. And that humanity--never before captured in guidelines or lists of therapeutic options--is what makes this project unique: It goes beyond mere knowledge, as important as that is, by entering the realm of human wisdom.

So I see the IBM-MSKCC collaboration as a way of bringing wisdom as well as knowledge to the care of cancer patients anywhere in the world. And that is the essence of why I see this project as one of profound importance.