Researchers led by Duc Hong Do from the Department of Cardiology at UCLA have developed an algorithm to detect the cause of cardiac arrest in a hospital setting.
In the United States, Over 200,000 patients suffer cardiac arrests while in the hospital every year, with only 30% surviving til discharge. The current approach to managing cardiac arrest, defined by basic life support (BLS) and advanced cardiac life support (ACLS), takes a one size fits all rather than account for all the different causes of cardiac arrest. This current approach was made for patients who suffer a cardiac arrest outside the hospital and have waited a prolonged period before entering the hospital. In a hospital setting, cardiac arrests are detected immediately and different protocols can be used to resuscitate the patient based upon the cause. Oftentimes, the standard ALS/ACLS procedures can cause further harm to the patient depending on the reason for the cardiac arrest.
Researchers led by Duc Hong Do from the Department of Cardiology at UCLA have developed an algorithm to detect the cause of cardiac arrest in a hospital setting. Based upon electrocardiogram (EKG) recordings their algorithm can predict the onset and cause of a cardiac arrest in real time. Additionally, this algorithm can provide cause-specific decision support for the management of cardiac arrest. Their invention can either display the cause of the cardiac arrest and subsequent treatment plan via a monitor or via audio through a speaker. Their results are based off a study of 89 patients, where they were able to identify 5 different types EKG patterns of cardiac with each type representing a different underlying cause.
heart attack, cardiac arrest, detection, diagnostic, treatment, machine learning, EKG