Research has been conducted within the Mayo Clinic to determine the added value of AI ECGs for earlier detection of problems with the heart's pump function. This condition, also known as low ejection fraction, appears to be able to be diagnosed much earlier with AI, when the patient does not yet have any immediate symptoms. Cardiologist Peter Noseworthy's research also shows that the use of AI-ECG technology can be used cost-effectively for preventive screening.
Declining heart pump function is treatable with appropriate medication. However, the condition is difficult to recognize, especially when patients do not yet have symptoms. In those cases, it is not common for physicians to refer patients for an echocardiogram or other diagnostic test to check ejection fraction.
Peventive AI ECG examination
With the study, published in Mayo Clinics Proceedings: Digital Health, cardiologist Noseworthy and Professor of Health Services Research at the Mayo Clinic Xiaoxi Yao, argue that preventive screening using AI-ECG technology, such as during a routine outpatient visit, can lead to heart failure being diagnosed earlier and treated at an earlier stage. Thus, the progression of the disease can be slowed or stopped and potentially higher medical costs avoided.
The researchers studied the economic impact of using the AI-ECG tool by using real-world information from 22,000 participants in the established EAGLE study and identifying which patients had a weak heart pump and which did not. They simulated longer-term disease progression and assigned values for patient health burden and the resulting impact on economic value.
“We categorized patients as either AI-ECG-positive, meaning we would recommend further testing for low ejection fraction, or AI-ECG-negative with no need for further testing. Then we followed the normal care pathway and looked at what that would cost. Did they have an echocardiogram? Do they stay healthy or do they develop heart failure later and need hospitalization? We looked at different scenarios, costs and patient outcomes,” said Dr. Xiaoxi Yao.
Leveraging AI tools and algorithms to diagnose heart disease faster and more accurately led to the development of AI-guided ECG software last year, by Utrecht UMC and Cordys Analytics. This supports healthcare providers in making decisions and minimizing hospitalizations and complications in patients with untreated heart disease. By harnessing the power of advanced deep learning techniques, this software recognizes subtle patterns and abnormalities in ECG data that traditional diagnostic methods often overlook.
Cost-effectiveness
According to the study, the cost-effectiveness of using AI-ECG was $27,858 per quality-adjusted life year - a measure of quality of life and life years. The program was especially cost-effective in outpatient settings, with a much lower cost-effectiveness of $1,651 per quality-adjusted life year.
“We know that earlier diagnosis can lead to better and more cost-effective treatment options. To achieve that, we have established a framework for AI evaluation and implementation. The next step is to find ways to streamline this process so that we can reduce the time and resources needed for such a rigorous evaluation,” Dr. Yao said.
