David Ouyang, MD, a research scientist and cardiologist, and his team at the Kaiser Permanente Northern California Division of Research have been awarded a $5 million special research grant from the American Heart Association. They will study how augmented intelligence (AI)—often referred to as artificial intelligence—may speed and improve screening and treatment for cardiovascular disease, the leading cause of death in the U.S.
“We’re incredibly excited to receive this award,” said Dr. Ouyang. “This award is validation of the foundational scientific work that our lab has been leading in leveraging AI for echocardiography and cardiology to improve patient care.
“Many AI projects stop just at a manuscript or pilot, while actual deployment in health care systems is often more challenging and requires the support of many stakeholders,” he added. “In order to truly impact patient care, we need to evaluate AI in clinical workflows, and this award helps support a clinical trial to evaluate AI to identify hidden diseases.”
The Permanente Medical Group is part of the AMA Health System Member Program, which provides enterprise solutions to equip leadership, physicians and care teams with resources to help drive the future of medicine.
New large-scale models trained on millions of echocardiographic videos can analyze multiple cardiac features and diagnoses simultaneously and generate detailed reports that may help physicians improve diagnosis for heart failure, valve disease and other cardiac abnormalities.
By applying advanced AI models to cardiac imaging, researchers hope to identify missed patterns and early indicators of disease, allowing physicians to intervene sooner and manage patients’ overall health even better.
In an interview with the AMA, Dr. Ouyang discussed how these technologies are being developed and tested, what role AI may play in expanding the diagnostic value of routine heart ultrasounds and what these innovations could mean for physicians and patients.
AMA: What aspect of your research led to this grant?
Dr. Ouyang: Our lab has been focused on the training of large AI models on large medical imaging datasets. In this grant, we focus on opportunistic screening. Imaging is one of the most common and rich forms of diagnostics that patients encounter, and the idea we are exploring is whether there is hidden information in the images that is not recognized in standard clinical workflows by physicians.
For example, echocardiograms or cardiac ultrasounds are interpreted by cardiologists but contain images of the liver that cardiologists are not trained to assess. By leveraging AI, we hope to get more value from these images. In this study, we will see if we can use heart imaging to diagnose liver disease earlier by getting patients to a specialist to assess their noncardiac issues.
AMA: What are the greatest challenges or opportunities you see for AI in advancing heart health?
Dr. Ouyang: There is a lot of information in electronic health records and medical images that can help diagnose patients at an earlier disease stage but can be missed in the standard of care. We think that AI has the potential to help us identify this critical information earlier.
However, we must first study how well these algorithms work in clinical practice and be able to demonstrate to stakeholders across the health care system that these technologies can be trusted.
By doing clinical trials we are generating evidence that can inform how we move AI from the research to the clinical setting.
AMA: How does this grant affect your next steps?
Dr. Ouyang: There are thousands of manuscripts being published on new AI technologies in health care, but very few of these papers actually look at how these AI technologies will impact patients in clinical practice. Our trial is designed to look at how the algorithms we are studying in the lab can be used in the clinic to study their impact on patient care. The results will help inform clinicians and patients about how AI can impact patient care.
AMA: Will the results of this study help guide physicians trying to do the same thing?
Dr. Ouyang: We previously designed one of the first and only blinded randomized clinical trials of AI in medicine. We hope our trial design and experience can help other physicians and researchers study how to implement AI in the clinic.
AMA: How do you envision AI tools changing the ways cardiologists diagnose or treat heart disease in the future?
Dr. Ouyang: I like to think of AI as a safety net, another set of eyes to augment physicians to improve quality of care. AI is not ready to be fully autonomous, like a self-driving car, but it can be helpful to receive thoughtful suggestions and assistance in many areas of clinical care.
AMA: What are the key takeaways around safety, accuracy and the promise of AI in cardiovascular medicine?
Dr. Ouyang: Our clinical trial is just getting underway and it will be ongoing until the end of 2028. This trial will allow us to assess whether AI screening of liver disease changes the disease trajectory in patients with earlier diagnosis of liver disease. By the end of the project, we hope to be able to report not only on the effectiveness of this AI technology but the steps that are needed to have this type of AI technology ready for widespread use in clinical care.
