Low blood pressure, the most common side effect of hemodialysis, can now be predicted with artificial intelligence (AI).
Seoul National University Hospital Nephrology Department Professor Han Seung-seok and Seoul National University Graduate School of Convergence Science and Technology Professor Kwak No-joon have developed a predictive model for hypotension during hemodialysis, according to the Clinical Journal of American Society of Nephrology.
The incidence of hypotension during hemodialysis is known to be about 20%, but it is difficult to predict. Most of them complain of nausea, vomiting, and convulsions and, in severe cases, lead to other problems, such as heart ischemia.
According to the professor’s team, it was difficult to predict hypotension due to frequent changes in blood pressure and various causes during dialysis.
The subject of this study is about 261,000 data from 9,292 hemodialysis patients. As a result of the analysis, a total of about 28,000 cases of hypotension occurred.
By learning this data, the research team created an artificial intelligence prediction model that predicts the occurrence of hypotension during hemodialysis within an hour. As a result of the test, it was found that the predictive ability was excellent.
According to the professor’s team, changing blood pressure data was the biggest factor in predicting performance. Therefore, in order to develop a model with high prediction accuracy, a vast amount of blood pressure data is required.
Professor Han said, “The 5-year mortality rate of hemodialysis patients is close to 40%, so hypotension during hemodialysis is most closely related to the risk of death.” It is the first step toward improving the quality of the product.”
Copyright © Medical Tribune Unauthorized reproduction and redistribution prohibited