AI identifies Atrial Fibrillation 92.9% from 1 lead wearable EKG/ECG

Atrial fibrillation commonly also known as AFib or AF is an irregular heartbeat (arrhythmia) that often increase the risk of blood clot, stroke,heart failure or other heart-related complications. It is the most common type of heart arrhythmia. With atrial fibrillation condition, the two upper chambers of the heart (the atria) beat chaotically and irregularly -out of coordination- with the two lower chambers of the heart (the ventricles). This could mean that the heart does not pump blood around the body as efficiently as it should and could lead to many complications. Some of the symptoms of Atrial fibrillation include a fluttering heartbeat, irregular pulse, weakness, tiredness, and dizziness. Atrial Fibrillation may occur in brief episodes, or it may be a permanent condition. If the doctors and medical professionals suspect that you have atrial fibrillation, they do conduct tests such as an electrocardiogram(ECG) or echocardiogram.

A study conducted and published in the year 2013 reveals that 33.5 million people around the world have AFib (Chugh et al., 2013). That is approximately 0.5 percent of the global population. Out of these numbers, 20.9 million were men and 12.6 million women. According to Cleveland Clinic, AFib affected about three million Americans in the year 2005, and the figure is projected to rise to eight million by the year 2050 (Pietrangelo, n.d.). There has been a progressive increase in the global prevalence and incidence of Atrial fibrillation with significant effects on associated morbidity and mortality. Such findings have implications for the public health policies and healthcare costs. The risk of Atrial Fibrillation increases with age, and in America, the median age for a woman with AFib is 74, and for men, it is 66 years.

According to the Centers for Disease Control and Prevention, an estimated 2.7 to 6.1 million Americans currently have AFib and with the aging of the United States population ("Atrial Fibrillation Fact Sheet|Data & Statistics|DHDSP|CDC," 2017). These numbers are expected to rise. Atrial fibrillation increases an individual’s risk for stroke by four to five times compared with stroke for people who do not have AFib. It causes 15%-20% of ischemic strokes. Ischemic strokes occur when the blood flow in artery to the brain is blocked. The brain depends on its arteries to bring fresh blood from the heart and lungs. If the artery remains blocked for more than a few minutes, the brain cells may die. Thirty-five percent of all AFib patients often have strokes. Atrial Fibrillation exacts a significant clinical burden. For instance, the condition is an independent predictor of mortality and is associated with more deaths, independent of other risk factors.

More than 750,000 hospitalizations occur each year because of AFib, and the condition contributes to an estimated 130,000 deaths each year. Atrial Fibrillation is a big issue in the United States since it costs the country approximately $6 billion annually, with the medical costs for people who have the condition being $8,705 higher per year than those who do not have AFib. The number of patients with Atrial Fibrillation in 2030 in Europe is estimated to be 14-17 million. AF increases the development of heart failure and adversely affects the quality of life, including cognitive function.

Diagnostic testing for AF may include an electrocardiogram (ECG) to check for the heart’s electrical activity. The most common symptoms include dyspnea, palpitations, chest pain, fatigue, and dizziness. Manual pulse palpation may determine the presence of irregular heartbeats that warrants further investigation using a 12-lead ECG. These tests are typically done with 12-lead ECG and require trained cardiologist and technician. Proficiency is a must. The correct placement of the electrodes is dependent on the accuracy of finding. Procedure can cause some discomfort and skin injury, especially on the elderly, if not performed carefully. (10) Equipment should be cleaned and disinfected at regular intervals to prevent infections (10). For operator and patient safety, peripheral equipment and accessories that can come in direct patient contact must be in compliance with all appropriate safety, EMC, and regulatory requirements (11).

Electrocardiographic results in AFib may include the absence of P waves, the presence of low amplitude and high frequency atrial fibrillary waves (Cantillon, 2014). Another test that may be conducted is the Holter monitor- a mobile ECG that can monitor the heart rhythms for several days. The medical professionals could look for heart abnormalities using an echocardiogram, which is a non-invasive test that produces images of the heart.

A preliminary study suggests a small device synced to a smartphone is currently used to help identify new cases of potentially deadly and irregular heart rhythm. Researchers in Hong Kong used this technology to check the feasibility of widespread community screening for Atrial Fibrillation. Over 13,000 individuals participated in the testing, and only 56 of those tested had results that could not be interpreted (Pallarito, 2016). Besides, over 100 were newly identified as having Atrial Fibrillation, and among them, 66 had no symptoms of AF. From these studies, it is fair to acknowledge that nearly half of the patients with a stroke caused by Arial Fibrillation have expressed no symptoms.

Several International studies also indicate that approximately 30% of patients with a stroke caused by AF did not know they have heart rhythm disorder until they have a stroke. AF can be better managed by monitoring patients for extended periods of time while they perform their normal daily activities and 1-Lead wearable ECG has allowed the patients to monitor their heartbeats from home. Some people do not know that they have AF because they do not have the symptoms and others may have one or more symptoms. The new technology allows people to test themselves for AF and detect irregular heartbeats while doing normal daily activities. These devices can be used for clinical diagnosis of AF by adding an ECG, and are a regular health monitoring connected to the internet via a smartphone (Iskandar, Kolla, Schilling & Voelker, 2016).

The ECG signal consists of low amplitude voltages in the presence of high offsets and noise. The signals can incorporate the latest diagnostic features and symptoms of AF, and with a trained cardiologist or technician, it is possible to diagnose AFib using the signal generated by the wearable devices. The resulting large amount of signal data need to be inspected second by second for any indications of problematic arrhythmias. These signals require precise, accurate and professional analysis by the medical doctors to determine any symptoms of AFib. The signal generated by these wearable devices are used in to extract vital characteristics, information or symptoms regarding AF or any abnormal heart activities.

New studies show that when the wearable devices are paired with an artificial intelligence-based algorithm or a neural network algorithm, they can detect severe and often symptomless heart arrhythmia, atrial fibrillation. The AI-based models developed based on Multi-layered neural networks and Deep learning by identifies 92.9% of patients with Afib, while the remaining 7.1% of the times correctly identified for irregular rhythm (which is a pre-condition for AFib) with abnormal conditions. The above artificial intelligence model also detects irregular heartbeats, cardiac abnormalities and ECG Morphology based parameters like (RR, PR, QRS, QTc intervals etc.). Unlike the 12-lead ECG method, these wearable devices do not require trained technicians to set-up. Afib occurs initially at random intervals, hence, requires periodic measurements or whenever individuals have feel of discomfort. Wearable devices are easy to do anytime, easy to use and small enough to carry while travelling. Patients use / wear portable devices paired with a mobile app and carry out their normal day-to-day activities without affecting their normal walk-of life while the device records each heartbeat for analysis (Dent, 2017).

AI-based Afib diagnosis presents an innovative opportunity to monitor, capture and prompt medical therapy for atrial fibrillation without any active effort from patients. While AI based Afib diagnosis technology screening cannot replace other more conventional monitoring methods, it has the potential to successfully screen those at an increased risk and lower the number of undiagnosed cases of AF (Rahman, 2017). If we are to provide these AI-based interpretations to Cardiologists, this will significantly increase their diagnosis accuracy and productivity as they are not only short supply but also overworked thereby reducing the error rates and cost associated with such conditions such as AF. Atrial Fibrillation has become the burden on health care cost not only in US but globally, as most patients remain asymptomatic and undiagnosed, thus leading to the raised mortality rate throughout the world. This mobile device is a major breakthrough in the health industry as its user friendliness makes it acceptance and use convenient to patients. It will allow self-care thus leading to early detection of the disease and improving the quality of life thereby decreasing the overall burden on healthcare cost.