MACHINE LEARNING IN DETECTION OF HEART ARRHYTHMIAS

Stanford University researchers have shown that a machine learning model can identify heart arrhythmias from electrocardiogram(ECG) better than an expert doctor. The automated approach is also significant because it could not make a quality care readily accessible in areas where resources are scares. So, first thing to know what is Heart arrythmias?

HEART ARRHYTHMIAS

Heart arrhythmias is a group of conditions in which heart beat is irregular, too fast or too slow. A heart rate that is too fast- 100 beats minute in adult is called Tachycardia & a heart rate that is too slow- below 60 beats per minute is called Bradycardia.

Cardiac arrhythmia occurs when electrical impulses in the heart don't work properly. There may be no symptoms. Alternatively, symptoms may include a fluttering in the chest, chest pain, fainting or dizziness. It may feel like a fluttering or racing heart & may be harmless. However, some heart arrhythmias may cause bothersome- sometimes even life threatening signs & symptoms.

Heart arrhythmias treatment can often control or eliminate fast, slow or irregular heartbeats. In addition, because troublesome heart arrhythmias are often made worse or are even caused by a weak or damaged heart. Most arrhythmias can be effectively treated. Treatments may include medications, medical procedures such as a pacemaker & surgery.

HOW MACHINE LEARNING HELPS IN HEART ARRHYTHMIAS

People suspected to have an arrhythmia will often get an electrocardiogram (ECG) in a doctor’s office. However, if an in-office ECG doesn’t reveal the problem, the doctor may prescribe the patient a wearable ECG that monitors the heart continuously for two weeks. The resulting hundreds of hours of data would then need to be inspected second by second for any indications of problematic arrhythmias, some of which are extremely difficult to differentiate from harmless heartbeat irregularities.

                Device used for detecting & monitoring heart arrhythmias using ML algorithm.

The research team, led by prof. Andrew Ng, trained their deep-learning algorithm on data collected from iRhythm’s wearable ECG monitor. They took approx. 30000, 30-second clips from various patients that represented a variety of arrhythmias. To test the accuracy of the algorithm, the researchers gave a group of three expert cardiologists 300 undiagnosed clips and asked them to reach a consensus about any arrhythmias present in the recordings. Working with these annotated clips, the algorithm could then predict how those cardiologists would label every second of other ECGs with which they were presented, in essence giving a diagnosis.

The group have six different cardiologists, working individually, diagnose the same 300 clip set. The researchers then compared which more closely matched the consensus opinion-the algorithm or the cardiologists working independently. They found that the algorithm is competitive with the cardiologists, and able to outperform cardiologists on most arrhythmias.

Here is the video showing how machine learning helps in detection of Heart arrhythmias :-

In addition to cardiologist-level accuracy, the algorithm has the advantage that it does not get fatigued and can make arrhythmia detections instantaneously and continuously.

Long term, this algorithm could be a step toward expert-level arrhythmia diagnosis for people who don’t have access to a cardiologist, as in many parts of the developing world and in other rural areas. More immediately, the algorithm could be part of a wearable device that at-risk people keep on at all times that would alert emergency services to potentially deadly heartbeat irregularities as they’re happening.