IBM’s Watson will use medical images to diagnose heart diseaseMar 5, 2017
Getting treatment for heart disease depends on a diagnosis from doctors, who can occasionally miss the subtle signs of trouble.
IBM thinks it can help those doctors through artificial intelligence — namely its Watson technology famous for besting Jeopardy champions and researching cancer. The company announced the introduction of its newest feature as part of its broader expansion of Watson Health’s medical imaging initiative, which will now include 24 healthcare organizations around the world.
This is a different challenge for Watson. For the first time, IBM’s technology will be looking over medical data that includes images such as ultrasounds, x-rays and other types of visuals used by medical professionals. Watson will first be employed in figuring out which patients need follow-up care for aortic stenosis, or AS.
This isn’t Watson’s first entry into healthcare, even outside its high-profile cancer research. The AI platform has found genes linked to ALS and worked to develop new drugs across diseases. The program has faced some setbacks, like recent mismanagement of Watson’s involvement in research at MD Anderson Cancer Center in Houston, Texas.
Watson’s new application in imaging will add to those fields. AS, the first imaging application for Watson, is a condition in which the heart’s aortic valve narrows and blocks blood flow to the rest of the body. It is one of the most difficult conditions for cardiologists to diagnose, said Jaime Murillo, a cardiology specialist at Sentara Healthcare.
Watson could help Murillo look at an image of a mass and figure out if it’s a tumor, an infection or just an anatomical quirk. The healthcare system in Virginia is one of IBM’s partners in developing imaging technology.
Watson Clinical Imaging Review will be the first cognitive imaging offering from IBM. The company has been working on developing cognitive imaging for over a decade, said Anne Le Grand, vice president of imaging for Watson Health.
“What we’re doing is looking retrospectively at medical records. We’re helping clinicians go back and look and say, ‘Are there patients we may want to bring back in?’ Le Grand said.
Watson will read cardiologists’ medical reports, remember data from other sources of medical information and analyze images of patients’ hearts. IBM tested the technology during a pilot study. The company declined to provide any numbers about how many patients were flagged for follow-up care by Watson other than to say it made a “big difference.”
Apart from identifying patients who might need follow-up care, Watson will look across patient populations to figure out similar patients who could benefit from follow-up visits, even if their ultrasounds or other images weren’t analyzed by Watson. Then, Watson will move into predictive care, helping to recommend treatment for patients who could be at risk based on the computer’s AI analysis.
“The response is twofold: improving the quality of diagnosis and the consistency of diagnosis,” Le Grand said.
Watson’s imaging technology has potential beyond heart disease including for breast cancer, for pulmonary and brain disease, and for ocular diseases like diabetic retinopathy. In the near future, IBM plans to expand Watson’s cardiovascular imaging work from AS to nine other cardiovascular conditions, including heart attacks, valve disorders, cardiomyopathy or disease of the heart muscle and deep vein thrombosis.
Murillo also sees potential for Watson in congestive heart failure and cardio-oncology, or for patients who need care for both heart disease and cancer. AS came first after Tanveer Syeda-Mahmood, chief scientist for medical sieve radiology at IBM Research, pursued the field for Watson’s first imaging application when her father was misdiagnosed over a decade ago.
The technology will be available to most U.S. healthcare practitioners later this year.
IBM is bullish on this technology. The company thinks it has major potential for practitioners across medical fields.
“Expect white papers,” Le Grand said.