Researchers at Stanford University in the United States have recently developed a deep learning algorithm that diagnoses 13 different types of arrhythmias by analyzing ECG data generated by wearable monitoring devices, even more accurately than cardiologists. This outcome can be used to improve the diagnosis and treatment of patients with arrhythmia in remote areas.
Potential arrhythmia patients usually go to the doctor and are examined by a doctor using an electrocardiograph. If the electrocardiograph does not detect a problem, the doctor may allow the potential patient to use the wearable device for two weeks of continuous monitoring of the heart rhythm. The time it takes for the device to generate data spans more than 300 hours, and doctors need to analyze the data for each second to find signs of arrhythmia. Hazardous heart rhythm data and endangered heart rhythm data are often extremely difficult to distinguish.
Stanford University’s news bulletin said that Wu Enda, the head of the school’s machine learning team and a famous artificial intelligence expert, found that this was a data issue. Researchers have developed a deep learning algorithm that can diagnose different types of arrhythmias based on ECG signals. Working with companies that provide wearable rhythm monitoring equipment, they acquired approximately 36,000 ECG data samples to train a deep neural network model. After 7 months, this neural network model is more accurate than a cardiologist in diagnosing arrhythmias, and in most cases even more than a doctor. Related research papers have been published on the online open database arXiv, which contains preprinted scientific literature.
According to the researchers, there are many types of arrhythmias, and the differences are subtle, but they have a great impact on how to deal with the arrhythmia found. For example, there are two types of arrhythmias known as secondary atrioventricular block, which look very similar, but one requires no treatment and the other requires immediate observation. Their research results not only to detect signs of arrhythmia, but also to detect different types of arrhythmia with high accuracy, which is unprecedented. In addition, the advantage of this algorithm is that it does not fatigue and can make an immediate diagnosis of arrhythmia.
Researchers hope that their algorithms will be used in the future to provide expert-level arrhythmia diagnosis for people who are unable to see a cardiologist in a remote or developing country. This algorithm can also be used with wearable rhythm monitoring devices for daily use by high-risk groups to notify emergency personnel when a potentially fatal heart rhythm is found.
Feed Additives Fruit Powder,Echinacea Purpurea Extract,Saccharomyces Cerevisiae Dextran,Polyphenols Echinacea Purpurea Extract
Fufeng Sinuote Biotechnology Co.,Ltd. , https://www.sntbiology.com