• Our healthcare system is heavily relied on imaging techniques to identify heart failures.
• Recent advances in AI can help detect heart failures more quickly.
• Mount Sinai researchers developed a special artificial intelligence (AI)-based computer algorithm that was able to learn how to detect subtle changes in electrocardiograms

In a typical year, about 1.2 million people go to emergency departments because they are experiencing shortness of breath, as per the report. Recently, 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.

When the heart pumps less blood than the body requires, heart failure, also known as congestive heart failure, occurs. For many years, doctors have relied heavily on an imaging technique known as an echocardiogram to determine whether or not a patient is suffering from heart failure. While beneficial, echocardiograms are time-consuming procedures.

However, recent advances in artificial intelligence suggest that electrocardiograms: a commonly used electrical recording device might be a quick and convenient alternative in these situations. Many studies have shown, for example, how a “deep-learning” algorithm can detect weakness in the left ventricle of the heart, which pushes freshly oxygenated blood out to the rest of the body.

AI comes to help

Mount Sinai researchers developed a special artificial intelligence (AI)-based computer algorithm that was able to learn how to detect subtle changes in electrocardiograms (also known as ECGs or EKGs) to predict whether a patient was suffering from heart failure.

Typically, an electrocardiogram involves a two-step process. Wire leads are taped to different parts of a patient’s chest and within minutes a specially designed, portable machine prints out a series of squiggly lines, or waveforms, representing the heart’s electrical activity. These machines can be found in most hospitals and ambulances and require minimal training to operate.

The research by Mount Sinai

The researchers at mount Sinai used a programmed computer to read patient electrocardiograms along with data extracted from written reports summarizing the results of corresponding echocardiograms taken from the same patients.  In this case, the written reports served as a baseline against which the computer could compare electrocardiogram data to learn how to detect weaker hearts.

Natural language processing programs helped the computer extract data from the written reports. Meanwhile, special neural networks capable of discovering patterns in images were incorporated to help the algorithm learn to recognize pumping strengths.

“The computer then read more than 700,000 electrocardiograms and echocardiogram reports collected from 150,000 Mount Sinai Health System patients from 2003 to 2020. Data from four hospitals was used to train the computer, whereas data from a fifth one was used to test how the algorithm would perform in a different experimental setting.”

The algorithm appeared to be effective at predicting which patients would have healthy or very weak left ventricles based on preliminary results. The left ventricle ejection fraction was used to determine strength, which is an estimate of how much fluid the ventricle pumps out with each beat as seen on echocardiograms. Healthy hearts have an ejection fraction of 50 percent or higher, whereas weak hearts have an ejection fraction of 40 percent or less.

The algorithm was 94% accurate in predicting which patients had a healthy ejection fraction and 85% accurate in predicting those with an ejection fraction of less than 40%.

These findings suggest that using an AI-ECG algorithm to identify patients at risk of repeat heart failure hospitalizations could provide a unique opportunity to prevent this.

In a nutshell

Finally, the additional analysis suggested that the algorithm may be effective at detecting heart weakness in all patients, regardless of race and gender. AI-enhanced ECGs are quicker and outperform current standard-of-care tests. The results suggest that high-risk cardiac patients can be identified quicker in the emergency department and provide an opportunity to link them early to appropriate cardiovascular care.

Want to learn more about artificial intelligence & its significance in the modern world from industry experts?

Reserve your time for Global Artificial Intelligence Summit & Awards Organized by AICRA happening on the 06th & 07th of October 2022.

Please click here to know more.