So researchers in the United Kingdom turned to machine learning to find a way to provide doctors with a fast and accurate way to diagnose heart attacks, helping to shorten the time needed to make a diagnosis and provide more efficient and effective treatment for patients, according to New Atlas, according to Nature Medicine.
One of the best and one of the best ways to diagnose a heart attack is to measure troponin levels in the blood. Troponin is released when heart muscle is damaged, and levels of the protein usually increase sharply within 3 to 12 hours after a heart attack, with a peak about 24 hours later.
Many hospitals around the world have adopted diagnostic pathways that include assessing troponin levels when a patient is suspected of having a heart attack.
However, there are some limitations. Measuring troponin requires timely collection of blood samples, which can be a challenge in emergency department work because ER doctors only classify patients as low, intermediate, or high risk of heart attack. Without considering other important information such as the time of onset of symptoms or ECG results, they also do not take into account the influence of sex, age and comorbidities.
machine learning algorithm
Now, British researchers have developed an AI-based machine learning algorithm that is fast and accurate. The algorithm is called CoDE-ACS, which stands for Collaborative Diagnosis and Evaluation of Acute Coronary Syndrome, and was designed to calculate the probability of a heart attack for an individual patient.
The researchers also used data from 10,286 patients with Bheart attacks Potential in 6 countries around the world. The machine learning algorithm was “taught” using a patient’s gender, age, ECG results and medical history, as well as troponin levels, to determine the likelihood of a heart attack.
99.6% accuracy rate
Compared to existing methods, the researchers found that CoDE-ACS can rule out a heart attack in more than twice as many patients, with an accuracy of 99.6%.
The algorithm also accurately predicted heart attack across subgroups, including men and women, the elderly, and those with renal impairment or those who presented to hospital early after the onset of symptoms.
According to the researchers, the innovative CoDE-ACS algorithm can prevent unnecessary hospital admissions for patients who are unlikely to have had a heart attack or those who are at low risk of heart muscle damage or death after a heart attack, which can make treatment in cases Emergencies are more efficient and effective as a result of the speed and accuracy of identifying patients, who can safely go home and which need to stay for further testing.
In the context, researcher Nicholas Mills said, “For patients who suffer from severe chest pain due to a heart attack, early diagnosis and treatment saves lives,” noting that there are “many health conditions that cause these common symptoms, and therefore the diagnosis is not clear in any case.” all cases”.
While he added, “Harnessing data and artificial intelligence to support clinical decisions has huge potential to improve patient care and efficiency in crowded emergency departments,” revealing that “the CoDE-ACS algorithm is currently being piloted in Scotland to see if it can reduce pressure on crowded emergency departments.”
#science. #fast #accurate #diagnose #heart #attack