Thereheart failure can be traced by a algorithm developed through the‘artificial intelligence and was also trained for recognize and detect subtle changes in the electrocardiograms (known as ECG or ECG): they declared it researchers from the Hasso Plattner Institute for Digital Health at Mount Sinai.
The study was published in the scientific journal Journal of the American College of Cardiology: Cardiovascular Imaging.
Heart failure diagnosed with AI: some research details
“We have shown that deep learning algorithms can recognize blood pumping problems on both sides of the heart from ECG waveform data “, he has declared Benjamin S. Glicksberg, Ph.D., assistant professor of genetics and genomic sciences, member of theHasso Plattner Institute for Digital Health at Mount Sinai and senior author of the study: “Usually, diagnosing this type of heart condition requires expensive and time-consuming procedures. We hope that this algorithm will allow a faster diagnosis of heart failure or heart failure ”.
The study was conducted by Akhil Vaid, MD, a researcher who works both in the laboratory Glicksberg than in the one led by Girish N. Nadkarni, MD, MPH, CPH, Associate Professor of Medicine at the Icahn School of Medicine at Mount Sinai, Head of the Division of Data-Driven and Digital Medicine (D3M) and senior author of the study.
Heart failure and congestive kidney failure affects an estimated 6.2 million Americans and occurs when the heart pumps less blood than the body normally needs. For years, doctors have relied heavily on an imaging technique called echocardiography to assess whether a patient might be suffering from heart failure.
While useful, echocardiograms can be labor-intensive procedures that are only offered in select hospitals.
However, Recent discoveries in artificial intelligence suggest that electrocardiograms, widely used electrical recording devices, could be a quick and readily available alternative in these cases. For example, many studies have shown that a “deep learning”Can detect weakness in the heart’s left ventricle, which pushes freshly oxygenated blood to the rest of the body.
In this study, the researchers described it development of an algorithm that not only evaluated the strength of the left ventricle, but also of the right ventricle, which takes deoxygenated blood flowing from the body and pumps it to the lungs.
“While attractive, it has traditionally been difficult for doctors to use ECG to diagnose heart failure. This is partly due to the fact that there are no established diagnostic criteria for these assessments and because some changes in ECG readings are simply too subtle to be detected by the eye. human”, said Dr. Nadkarni.
“This study represents an exciting step forward in the search for hidden information within ECG data that can lead to better screening and treatment paradigms using a relatively simple and widely available test.“, Continued the researcher.
Typically, an EKG involves a two-step process. The cables are attached to different parts of the patient’s chest and within minutes a specially designed portable machine prints a series of wavy lines, or waveforms, representing the electrical activity of the heart.
These machines can be found in most hospitals and ambulances in the United States and require minimal training to operate.
For this study, the researchers programmed a computer to read the patients’ electrocardiograms along with data extracted from written reports summarizing the results of the corresponding echocardiograms taken from the same patients. In this situation, the written reports served as a standard data set for the computer to compare with the electrocardiogram data and learn to identify the weakest hearts.
Natural language processing programs helped the computer extract data from written reports. Meanwhile, special neural networks capable of discovering patterns in images have been incorporated to help the algorithm learn to recognize pumping forces.
“We wanted to promote the state of the art by developing an artificial intelligence capable of understanding the whole heart in a simple and economical way “, said Dr. Vaid.
The computer then read more than 700,000 electrocardiograms and echocardiograms obtained from 150,000 patients of the Mount Sinai Health System from 2003 to 2020. Data from four hospitals was used to train the computer, while data from a fifth was used to test the algorithm’s performance in a different experimental setting.
“A potential benefit of this study is that it involved one of the largest ECG collections from one of the most diverse patient populations in the world “, said Dr. Nadkarni.
Initial results suggested that the algorithm was effective in predicting which patients would have healthy or very weak left ventricles. Here strength was defined by left ventricular ejection fraction, 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 greater, while weak hearts have those equal to or less than 40 percent.
The algorithm was 94% accurate in predicting which patients had healthy ejection fraction and 87% accurate in predicting those who had less than 40% ejection fraction. The algorithm was 94% accurate in predicting which patients had healthy ejection fraction and 87% accurate in predicting those who had less than 40% ejection fraction.
However, the algorithm was not as effective in predicting which patients would have a slightly weakened heart. In this case, the program was 73% accurate in predicting patients who had an ejection fraction between 40 and 50%. Further results suggested that the algorithm also learned to detect right valve weak spots from electrocardiograms. In this case, the weakness was defined by more descriptive terms extracted from the echocardiogram reports. Here the algorithm was 84% accurate in predicting which patients had weak right valves.
“Our results suggested that this algorithm could eventually help doctors correctly diagnose failure on both sides of the heart“Said Dr. Vaid.
Finally, further analysis suggested that the algorithm could be effective in detecting heart failure in all patients, regardless of race and gender: “Our results suggest that this algorithm could be a useful tool to help doctors combat heart failure suffered by a variety of patients”added Dr. Glicksberg. “We are carefully designing potential studies to test their effectiveness in a more real-world environment“.
#Heart #failure #artificial #intelligence #detects