Utilizing AI May Help in Diagnosing Heart Failure

With heart failure costs rising, such AI interventions may be crucial

AI enhanced electrocardiograms (ECG) may be able to accurately detect heart failure in patients being evaluated in the ER for shortness of breath, according to a study published in Circulation: Arrythmia and Electrophysiology.

The method could detect decreased heart function more accurately and quickly than standard blood tests, the study found.

In a typical year, about 1.2 million people go to emergency departments because they are experiencing shortness of breath, researchers stated. This year, those numbers are much higher because difficulty breathing is a major sign of COVID-19. When providers suspect a patient is having heart problems, they usually perform an ECG, a 10-second recording of the heart’s electrical activity.

“An abnormal ECG raises concern about underlying cardiac abnormalities but are not specific for heart failure,” said Demilade Adedinsewo, MD, MPH, lead author of the study and chief fellow in the division of cardiovascular medicine at Mayo Clinic in Jacksonville, Florida.

ED physicians also rely on blood levels of natriuretic peptides. When heart failure is present, these biomarkers are elevated in the blood, but these biomarkers are also influenced by age, obesity, kidney disease, severe infection, pulmonary hypertension, and abnormal heart rhythms.

To develop the artificial intelligence-enabled ECG, researchers used data on thousands of patients to train computers to distinguish between the ECG patterns of people ultimately diagnosed with left ventricular systolic dysfunction (LVSD) and those without LVSD.

In about ten seconds, standard ECG recordings can be analyzed with the resulting AI software application to identify likely LVSD.

The team tested the accuracy of AI-enabled ECG to identify LVSD in ED patients with shortness of breath compared to the results of biomarker blood tests. Researchers applied the AI tool to the ECGs of 1,606 patients who had received an ECG and blood testing in the emergency department, later followed by definitive testing using an echocardiogram.

Researchers found that AI-enhanced ECG was better than standard blood tests in identifying which patients have severe LVSD, with a performance measure of 0.89 versus 0.80. The AI-enhanced ECG also achieved a performance measure of 0.85 when identifying patients with less severe but abnormally low pumping ability.

The results also showed that while several factors can influence blood test results, AI-enhanced ECG performed just as well in women and men and among patients in different age groups.

The results have important implications for heart failure identification and care, researchers noted. The team stated that in the U.S., healthcare costs for patients with heart failure are projected to increase from $20.9 billion in 2012 to $53.1 billion in 2030.

It’s critical that patients experiencing heart failure are accurately and quickly identified in the ED, and that appropriate therapies are initiated early to potentially reduce costs and readmissions.

“We observed that patients with LVSD were significantly more likely to be rehospitalized with heart failure,” researchers said.

“These findings suggest that the use of an AI-ECG algorithm could potentially identify patients at risk for repeat heart failure hospitalizations and provides a unique opportunity to implement specific interventions to prevent this while in the ED, including early follow up with a cardiologist, initiation of guideline directed medical therapy, and social work if needed.”

SOURCE: Health IT Analytics

About The Author