We are searching data for your request:
Upon completion, a link will appear to access the found materials.
Improved diagnosis of heart attacks could save lives
A newly developed test could help diagnose heart attacks faster and more effectively in the future. A computer algorithm, in combination with a blood test, determines whether people have had a heart attack or how high their risk of developing a heart attack is within the next 30 days.
A recent study by Christchurch Hospital and the University of Edinburgh has now found that a computer algorithm combined with a blood test can significantly improve the diagnosis of heart attacks. The results of the study were published in the English-language journal "Circulation".
Improved treatment by AI
The researchers have developed a new computer algorithm that, in combination with a blood test, determines whether someone has suffered a heart attack. This algorithm showed improved diagnostic speed and accuracy. A cardiogram and a blood test must still be carried out on those affected and the family history and symptoms often play an important role in the early diagnosis.
What can the algorithm determine?
The new algorithm only uses objective information such as age and gender and a blood test. In this way, he is able to eliminate distortions in the diagnosis, explains the research team. If patient characteristics were taken into account and a blood test was performed, a computer could determine the likelihood of heart attacks for patients as low or high. The algorithm is also able to predict the risk of a heart attack within the next 30 days relatively accurately, the researchers report.
Diagnosis is much more targeted
The results of the study showed that the computer algorithm was specially trained to estimate the personal risk of an individual patient, which made a much more targeted diagnosis possible, the research group explains further. The data from over 11,000 people around the world were evaluated for the study. This enabled the development of the new computer algorithm that could significantly improve the diagnosis of heart attacks in the future. The researchers believe that this is a major step in future healthcare. (as)
Author and source information
This text corresponds to the requirements of the medical literature, medical guidelines and current studies and has been checked by medical doctors.
Swell:
- Martin P. Than, John W. Pickering, Yader Sandoval, Anoop S.V. Shah, Athanasios Tsanas et al .: Machine Learning to Predict the Likelihood of Acute Myocardial Infarction, in Circulation (Query: 11.09.2019), Circulation
