Leveraging Machine Learning Techniques for Cardiovascular Care
In This Video
- Researchers at Massachusetts General Hospital and MIT are collaborating to integrate computer science with clinical care for cardiovascular disease
- Machine learning methods and artificial intelligence can help providers identify and manage high-risk patient subgroups
- Using neural networks, researchers have identified patients at risk for cardiovascular death
In this video, Collin M. Stultz, MD, PhD, cardiologist in the Cardiology Division at Massachusetts General Hospital and professor of Electrical Engineering and Computer Science at the Institute for Medical Engineering and Science at MIT, discusses his research integrating computer science and patient care for cardiovascular disease. Applying machine learning and artificial intelligence methods can identify high-risk patient subgroups and help providers manage those patients.
Subscribe to the latest updates from Cardiovascular Advances in Motion
Transcript
Computer scientists are interested in developing methods, and clinicians are interested in taking care of their patients. What is really missing in the field are individuals who can bridge this divide, who can use, develop and apply sophisticated machine learning techniques to problems that really are at the heart of the care for patients with cardiovascular disease. It requires not only what techniques to use, but because the questions are very specific in the cardiovascular sphere, it also involves developing new methods that can inform physicians who take care of patients with cardiovascular disease.
What is essential to this is not just the application of computer science methods, but it's the merging of expertise, both in computer science and in cardiovascular medicine. So, integrating the various knowledge that's available at Massachusetts General Hospital and MIT to create a format where clinicians and computer scientists can interact to address important questions that are essential to the care of patients with cardiovascular disease.
When we first began our work at MIT, we did a lot of very basic science research looking at the shapes of proteins and DNA and how that related to disease. That work is very important, but it probably would be hard for that to have an impact, a real impact on patients in my lifetime. What motivated us to move into this sphere is to see how we could use our knowledge about computation to develop tools that would help patients in the near-term. And a natural place to go was an application of machine learning techniques to problems in cardiovascular care.
So, we've developed a variety of different techniques over the years for processing physiologic signals. And a number of different machine learning methods for identifying high-risk patient subgroups. We can identify patients at risk for cardiovascular death using things like neural networks and other types of methods grounded in machine learning and artificial intelligence to determine how to manage these patients. So, where we'd like to see this is in five years, the establishment of some integrated center where Mass General and the Department of Computer Science at MIT work in unison and in conjunction to develop these techniques that really are of clinical use.
Learn more about the Corrigan Minehan Heart Center
Refer a patient to the Corrigan Minehan Heart Center