The number of cases of heart attack and stroke among adults has more than tripled since 1980, the Centers for Disease Control and Prevention reports.
That’s due to a variety of factors, including obesity, a lack of exercise, and poor lifestyle choices, according to a report from the Centers For Disease Control.
But a new algorithm developed by the University of Texas at Austin and funded by the National Institutes of Health (NIH) has the potential to help prevent heart attacks and stroke.
The algorithm, called “heart valves optimality,” or HOVE, uses machine learning algorithms to develop personalized algorithms that can improve the effectiveness of a heart valve’s function.
The algorithms, which were created by researchers at UT Austin and the University at Buffalo, are based on a system called “flow dynamics” that’s used in medical imaging.
In flow dynamics, scientists use computer models to simulate how fluid flows across the heart, allowing them to predict how it will behave when the heart is pumping.
The HOVE algorithm uses that flow dynamics model to predict what heart valve functions will change based on various parameters.
HOVE uses a computer model to create customized algorithms that optimize the functions of the heart valve, according a statement from the researchers.
The new algorithm, developed by a team at the UT Austin Center for Computer Vision, will be the first algorithm to use computer vision to develop an algorithm that could reduce the likelihood of heart valve malfunction and improve its function.
According to the study, the algorithms developed by this team are better than the algorithms currently used to identify heart valve defects.
In addition, the algorithm developed at UT UT Austin has been validated by other researchers, including the University College London, who have used it to predict heart valve damage in patients.
“Our research demonstrates the potential for a new generation of algorithms to improve heart valve functioning and predict disease, which is crucial to prevent heart attack,” said University of Tennessee’s J.P. Pritzker, chair of the department of mechanical engineering and a co-author of the study.
“The HOVE algorithms are one step toward an approach that can help prevent and improve heart disease.”
HOVE works by simulating the flow of fluid through a heart as it pumps in and out of a valve.
The simulation takes into account various variables, such as the length of time the heart has been pumping and the strength of the muscles around the valve, the researchers said.
These variables determine the type of fluid that flows through the valve and how fast it flows.
A new algorithm by UT Austin will be able to predict changes in valve function, and therefore prevent heart valve problems, the study said.
According the researchers, the new algorithm can be used to predict valve failure, and even identify heart valves that are failing at a much higher rate.
“We can predict the heart’s valve function at a level that will be more accurate than the human ability to perform the function,” said the study’s senior author, Dr. Tania L. Almeida.
“This new algorithm will allow us to predict the valve function in real time and intervene quickly if a valve is being underperforming.”
In the future, the authors hope that the HOVE model could be applied to other heart valves, such a ventricle valve, to see if it can improve valve function and prevent heart disease.
“A valve’s ability to pump and release blood depends on the number of pumping chambers that are present, the length and strength of muscles in the valve’s system, and the position of the valves valves head,” the researchers wrote.
“In this paper, we show that the combination of these factors, along with other parameters such as oxygen content and pressure, can be applied and predict the flow dynamics of a large-scale valve.”
Researchers hope that these algorithms can be developed to predict future heart valve disorders, such heart failure and valve failure due to inflammation, in the future.
“While heart valve dysfunction is a growing public health concern, it is difficult to pinpoint the cause and develop effective therapies,” the authors of the new study said in a statement.
“By developing a computer algorithm to predict disease progression, we hope to reduce morbidity and mortality, while improving patient care.”
The study was published in the Proceedings of the National Academy of Sciences.
Image Credit: UT Austin