Heart failure is a serious disease that affects millions of people around the world. This happens when the heart is unable to pump blood properly through the body. In advanced stages, it can be fatal. However, the biggest challenge is its early detection. Doctors face a big problem: Many patients with advanced heart failure are diagnosed late, which prevents them from getting timely treatment. Now, new research has raised hopes that Artificial Intelligence (AI) can change this situation.
What technology is being used now?
The study, conducted by scientists at Weill Cornell Medicine, Cornell Tech, Columbia University and NewYork-Presbyterian, was published in the journal npj Digital Medicine. Currently, doctors use a special test called a cardiopulmonary exercise test to diagnose advanced heart failure. This test measures heart and lung function, but it requires special equipment and trained staff, which are available only in large hospitals. Therefore, many patients do not receive this diagnosis.
What was found in the research
In a new study, scientists have created an AI system that can predict the severity of the disease by analyzing cardiac ultrasound (echocardiography) and the patient’s general medical history. The special thing is that this technology can estimate important parameters like VO2 max, which is generally measured only through cardiopulmonary exercise testing (CPET).
Using data of 1,000 heart failure patients
The system was built using data from nearly 1,000 heart failure patients, including ultrasound video, blood flow and heart valve activity. It was then tested on 127 new patients, where the AI model identified high-risk patients with about 85% accuracy. Researchers say that this technology can be easily used in everyday medical needs, because it uses already existing data. This will enable identification of patients who were previously unnoticed. However, scientists also believe that further testing is needed before this technology can be implemented on a large scale. It is equally important for the AI system to be reliable and ensure security.
