-1752819012997.webp)
According to World Heart Health Report 2023, 50 crore people around the world suffer from cardiovascular diseases, and two-crore lose their lives to the disease each year, and the numbers, over the years, have only grown as the world increasingly leans towards sedentary lifestyle. Most of these deaths though, happen because early diagnoses are not feasible. Structural heart disease, including valve disease, Congenital Heart Disease (CHD), and other issues that impair heart function, affecting millions of people worldwide. And, these conditions often remain undetected due to lack of a routine and affordable screening test. A common electrocardiogram (ECG) doesn’t normally detect structural abnormalities and ultra sound is an expensive deal for most. But, what if there was a way to make the common ECG much more efficient, and able to detect structural heart disease early? Well, according to new research, AI might just have made that possible.
-1752820162044.jpg)
What Is EchoNext?
The paper published in the journal Nature claimed that scientists had developed combined AI and common ECG to design EchoNext that identifies patients who should have an ultrasound (echocardiogram), a procedure used to diagnose structural heart issues.
Pierre Elias, an assistant professor of medicine and biomedical informatics at Columbia University Vagelos College of Physicians and Surgeons, who also led the study, stated, "We have colonoscopies and mammograms, but we have no equivalents for most forms of heart disease." EchoNext was created to analyse routine ECG data and identify when a cardiac ultrasound follow-up is necessary. "It identifies illnesses that cardiologists are unable to identify from an ECG. The researcher continued. EchoNext basically uses the cheaper test to figure out who needs the more expensive ultrasound,” he added.
EchoNext Outperforms Expert Cardiologists
EchoNext was created to evaluate routine ECG data and identify when a cardiac ultrasound follow-up is necessary. Over 12 lakh ECG pairs from 230,000 patients were used to train the deep learning model.
The screening tool showed high accuracy in identifying structural heart problems, such as heart failure due to cardiomyopathy, valve disease, pulmonary hypertension, and severe thickening of the heart, in a validation study involving four top hospital systems in the US. EchoNext correctly detected 77% of structural heart issues in a head-to-head comparison with 13 cardiologists on 3,200 ECGs. Cardiologists, on the other hand, had a 64% accuracy rate when diagnosing using the ECG data.
-1752820181755.jpg)
Understanding Structural Heart Disease
Conditions affecting the valves, walls, chambers, or muscles of your heart are referred to as structural heart disease. Cardiomyopathy, congenital heart disease, and heart valve disease are some of the prominent conditions recognised as structural heart disease . These can be inherited or they may develop over time. Your cardiologist should keep a close eye on you even if you don't always require treatment.
Also Read: How To Differentiate Between Heartburn And A Heart Attack? Know From Expert
Bottomline
The research team tested EchoNext on almost 85,000 patients undergoing ECG who had never had an echocardiogram before to see how well the tool performed in practice. More than 7,500 people, or 9% of the population, were classified by the AI tool as having a high risk of undiagnosed structural heart disease. Fifty-five percent of those identified by EchoNext as high-risk underwent their first echocardiogram. Almost three-quarters of them received a structural heart disease diagnosis, which is twice as many positive results as if everyone had their first echocardiogram without AI. The results highlight an obvious pattern. With the help of AI, medical science could be capable of saving more lives than ever before as the most threatening aspect of structural cardiovascular diseases is that they often remain elusive.