|Heart failure is a major healthcare concern across the world. The American Heart Association’s 2017 Heart Disease and Stroke Statistics Update revealed that the number of people diagnosed with heart failure is increasing / Photo by: Katarzyna Białasiewicz via 123RF|
Heart failure is a major healthcare concern across the world. The American Heart Association’s 2017 Heart Disease and Stroke Statistics Update revealed that the number of people diagnosed with heart failure is increasing. It is projected to rise by 46% by 2030, resulting in more than eight million people with heart failure. Several risk factors can cause heart failure, including an unhealthy lifestyle, smoking tobacco, and conditions such as diabetes and high blood pressure.
Paul Muntner, Ph.D., a professor and vice-chair in the Department of Epidemiology at the University of Alabama at Birmingham, stated that the increase in heart failures can be attributed to medical advances. This is because more and more people are surviving heart attacks, thus, facing higher heart failure risk afterward.
“The epidemics of diabetes and obesity both contribute to the rising number of patients who acquire heart failure -- our growing population of the elderly are particularly susceptible,” Mariell Jessup, M.D., a heart failure expert and former president of the American Heart Association, said.
Unfortunately, diagnosing heart failure can be difficult because its symptoms and signs commonly overlap with other conditions. While chest X-rays can be useful in identifying evidence of heart failure, a normal result doesn’t usually rule out a diagnosis of heart failure. The same situation can happen with an electrocardiogram (ECG) and natriuretic peptides. Thus, experts have suggested that a new clinical-decision rule (CDR) could help medical professionals achieve a more timely and accurate diagnosis of heart failure.
Evidence has shown a rampant rise in misdiagnosis of heart failures. A 2010 study published in European Cardiology Review (ECR) revealed that only 50% of patients with a clinical label of heart failure had confirmed diagnosis after formal assessment according to diagnostic criteria. Aside from that, there are some possible barriers to accurate diagnosis and effective management of heart failure. This includes limited diagnostic provisions, lack of confidence of primary care physicians in establishing an accurate diagnosis, concerns about the use of polypharmacy in older, and more.
Fortunately, artificial intelligence can help.
Detecting Heart Failure Through AI
It’s extremely important to have early and accurate diagnosis and treatment of heart failure because it can improve the quality and length of life for people suffering from the condition. A 2019 study conducted by researchers from the University of Surrey developed a neural network approach in accurately identifying congestive heart failure (CHF) with 100% accuracy. CHF is associated with high prevalence, significant mortality rates, and sustained healthcare costs.
According to Science Daily, an American website that aggregates press releases and publishes lightly edited press releases about science, the study published in the Biomedical Signal Processing and Control Journal had drastically improved the existing CHF detection methods. These methods are commonly focused on heart rate variability, which is both time-consuming and prone to errors. At the same time, the new model can deliver a 100% accuracy through a combination of advanced signal processing and machine learning tools on raw ECG signals.
Dr. Sebastiano Massaro, Associate Professor of Organisational Neuroscience at the University of Surrey, stated that the CNN model was trained and tested on large publicly available ECG datasets. “Our model delivered 100% accuracy: by checking just one heartbeat we are able to detect whether or not a person has heart failure. Our model is also one of the first known to be able to identify the ECG' s morphological features specifically associated with the severity of the condition,” Dr. Massaro added.
According to Forbes, a global media company focusing on business, investing, technology, entrepreneurship, leadership, and lifestyle, the researchers focused on detecting heart failure using five-minute recordings instead of 24-hours recordings. “This aspect offers a valuable potential for prospects of rapid interventions; nonetheless it is also important to keep in mind that we are talking about severe CHF patients only at the moment,” Dr. Massaro said.
Predicting Life Expectancy in Heart Failure Through AI
AI not only detects heart failure with 100% accuracy but also predicts life expectancy in heart failure patients. Last November 2019, cardiologists and physicists from the University of California - San Diego introduced a machine learning algorithm, which they developed for struggling heart failure patients.
Avi Yagil, professor of physics at the University of California - San Diego, teamed up with Eric Adler, cardiologist and director of cardiac transplant and mechanical circulatory support, and Barry Greenberg, professor of medicine at UC San Diego School of Medicine, to develop the AI tool. According to Health Europa, an online site that provides news and developments from across the entire spectrum of health care, the researchers developed an AI model that can predict life expectancy with 88% accuracy. At the same time, it can perform substantially better than other popular published models.
The researchers used de-identified electronic health record data of 5,822 hospitalized or ambulatory patients with heart failure at UC San Diego Health to develop the AI tool. They also identified eight variables collected for the majority of patients with heart failure, including diastolic blood pressure and red blood cell distribution.
"This tool gives us insight, for example, on the probability that a given patient will die from heart failure in the next three months or a year. This is incredibly valuable. It allows us to make informed decisions based on a proven methodology and not have to look into a crystal ball,” Adler said.
These new AI tools/methods are now included in the existing AI-driven diagnostic tools, promising a revolution in clinical approaches to evaluating medical data. All of these can significantly help the industry in the long run.
|Last November 2019, cardiologists and physicists from the University of California - San Diego introduced a machine learning algorithm, which they developed for struggling heart failure patients / Photo by: Andriy Popov via 123RF|