IBM’s Jeopardy-champion computer, Watson, has huge potential for helping physicians and other clinicians work with patients.
The leap from TV game show to physicians’ offices will probably take at least two years. But Watson’s understanding of natural language, vast storehouse of information and ability to keep up with rapidly changing medical research could significantly improve medical care.
The medical faculty at Columbia University and University of Maryland are helping program a Watson-type computer to assist clinicians.
A few years from now, consulting Watson could become a routine part of a clinician’s practice. Caregivers have traditionally resisted computerized assistance in diagnosis and treatment because the technology has been awkward to use and questionnaire-based systems have been too rigid. But Watson can “understand” descriptions of a patient’s symptoms in natural language, and it can even scan years of medical records and doctors’ notes to determine what diagnostic and therapeutic options it might suggest. Doctors can ask it questions using the same terms they would use in an e-mail to a colleague.
The amount of medical knowledge is doubling every five to seven years, according to various estimates. It’s impossible for anyone to keep up, but Watson could. It could also review the medical histories that patients accumulate as they age. In the next few years when genetic tests for all become commonplace, Watson would be able to analyze them, too. That will be challenging for most primary care physicians.
One quarter of all medical errors involve misdiagnosis or delayed diagnosis. For example, it often takes over 10 years before a patient is correctly diagnosed with celiac disease, which can cause malnutrition, stomach pain and constant diarrhea because the patient can’t digest gluten. It takes an average of four years to diagnose multiple sclerosis. Doctors fail to suspect these diseases because they have many symptoms that suggest more common problems.
Watson, by using the same capability it used on Jeopardy, can search its vast memory for a likely diagnosis. Then it may say that the evidence it has points with 88% confidence to the patient having dermatitis, but that there’s also a 6% likelihood she has celiac disease, something the doctor might not have even considered.
As is demonstrated weekly on TV shows like “House,” correctly reaching a rare diagnosis may be an exciting, glamorous aspect of medicine. But in practice, figuring out even common diagnoses with increased confidence, accurately and quickly, can keep scores of patients more healthy, getting them on treatment paths sooner.
Prescribing treatment is difficult because “one size doesn’t fit all”. A clinician doesn’t need Watson to treat high blood pressure. But Watson might help if the patient is an 80-year-old diabetic with prostate problems, because he may not be able to take certain medications. Watson could backstop a doctor with suggestions and warnings.
Most clinicians believe in practicing evidence-based medicine, which is a central theme of recent changes to U.S. health care laws. But it’s very difficult for a caregiver to extrapolate the results of medical research to the special case of an individual. Watson could go through FDA records of adverse reactions to drugs to see if there were reasons a patient shouldn’t get a common prescription.
Because Watson understands what people mean when they write in ordinary English, it could be trained to be an omnivorous reader of all sorts of both standard and unorthodox medical literature. The information to inform a clinician’s choices is out there somewhere.
Watson’s ability to hit all those sources of information could one day make it an indispensable aid to caregivers.
This year marks IBM’s Centennial – and Watson is another example of a long history of healthcare innovations, including the 1981 invention of an excimer laser technique that made photorefractive (LASIK) eye surgery possible. IBM also played a major role in developing the heart lung machine in the 1950s, and invented the first continuous blood separator in the 1960s, used to treat critically ill leukemia patients.
Dan Pelino is general manager for the IBM Healthcare and Life Sciences business. He leads the corporation in helping clients create smarter, more connected health care systems, working closely with public and private health care providers and payers, biotech and pharmaceutical companies, and medical device and instrument companies. He is an expert in the areas of health care transformation and health IT. He has advised numerous countries and states on health care IT-related issues. Under his leadership, IBM has been involved in helping transform and digitize health systems worldwide.