September is atrial fibrillation awareness month

Atrial fibrillation, an increasing epidemic

‍
 Atrial fibrillation (AF) is the most common type of arrhythmia. It affects over 33 million people in the world¹. Individuals with AF are 5 times more likely to suffer a stroke.². It has been estimated that 20-30% of stroke cases are caused by atrial fibrillation.

This type of arrhythmia consists of uncoordinated and disorganized beats that originate in a part of the heart called the atrium, which produces a rapid and irregular heart rhythm.

Healthy people with no other medical problems can develop AF. However, in most cases, there are some risk factors such as hypertension, diabetes, alcohol and other drug abuse, stress, obesity or a sedentary lifestyle¹.‍

How does atrial fibrillation happen?

Under normal conditions, our heart rate should be regular, whether we are at rest or doing physical activity. When atrial fibrillation occurs, the heart rhythm is no longer regular and steady. The heart "fibrillates" as if it were trembling. This loss of synchrony in the heart rhythm can lead to the formation of blood clots inside the heart triggering thrombus formation and stroke. Therefore, early detection of AF and diagnosis is key to start anticoagulation therapy to prevent stroke.

AF is strongly associated with age: it generally affects people 40 years old and older³. Since the aging population is increasing, AF prevalence has increased over 33% during the last 20 years⁴.‍

How is atrial fibrillation detected?

Some AF patients present symptoms that affect their daily life such as heart palpitations, dizziness, shortness of breath, and tiredness. However, other patients are completely asymptomatic. It is still not known why some patients present symptoms while others do not. Thus, the electrocardiogram (ECG), a clinical test to check heart rhythm, is needed to confirm an AF diagnosis.

In some patients, atrial fibrillation occurs only episodically, with crises alternating with periods of normal heart rhythm. In these cases, the arrhythmia may not be easily detected. In response to this problem, continuous heart rhythm monitoring is now possible nowadays thanks to sophisticated devices such as implantable cardiac monitors (ICMs), or external devices such as Holter monitors or wearables (fit bands, watches, textile adhesives, etc)⁵. 

How can AI help in AF detection? 

Remote continuous patient monitoring produces a massive amount of ECG data that must be reviewed by cardiologists. In recent years, artificial intelligence (AI) algorithms have been suggested as a potential solution for helping cardiologists with ECG analysis and cardiac abnormalities triage^6–11. Idoven’s AI-platform, known as Willem^TM (based in Madrid, Spain), is a powerful and scalable AI-platform for assisting cardiologists in ECG interpretation and arrhythmias diagnosis. Idoven has recently demonstrated that Willem^TM can improve AF episode detection from ICMs with a high accuracy of 91,32% and reduce false positive AF detections by 94%¹². Thus, AI solutions will reduce cardiologist’s workload enhancing patient care and prognosis, which results in improved survival rates for AF patients.

‍

_{1. Alshehri, A. M. Stroke in atrial fibrillation: Review of risk stratification and preventive therapy. Journal of Family and Community Medicine 26, 92–97 (2019).}

_{2. Marini, C. et al. Contribution of atrial fibrillation to incidence and outcome of ischemic stroke: Results from a population-based study. Stroke 36, 1115–1119 (2005).}

_{3. Lloyd-Jones, D. M. et al. Lifetime risk for development of atrial fibrillation: The framingham heart study. Circulation 110, 1042–1046 (2004).}

_{4. Lippi, G., Sanchis-Gomar, F. & Cervellin, G. Global epidemiology of atrial fibrillation: An increasing epidemic and public health challenge. Int J Stroke 16, 217–221 (2020).}

_{5. Svennberg E et al. How to use digital devices to detect and manage arrhythmias: an EHRA practical guide. Europace (2022).}

_{6. Hao, W. & Jingsu, K. Investigating Deep Learning Benchmarks for Electrocardiography Signal Processing. Arxiv (2022).}

_{7. Bollepalli, S. C. et al. Real-time arrhythmia detection using hybrid convolutional neural networks. J Am Heart Assoc 10, e023222(2021).}

_{8. Kwon, J. M. et al. Artificial Intelligence-Enhanced Smartwatch ECG for Heart Failure-Reduced Ejection Fraction Detection by Generating 12-Lead ECG. Diagnostics 12, 654 (2022).}

_{9. Zhang, J. et al. ECG-based multi-class arrhythmia detection using spatio-temporal attention-based convolutional recurrent neural network. Artif Intell Med 106, 101856 (2020).}

_{10. Attia, Z. I., Kapa, S., Noseworthy, P. A., Lopez-Jimenez, F. & Friedman, P. A. Artificial Intelligence ECG to Detect Left Ventricular Dysfunction in COVID-19: A Case Series. Mayo Clin Proc 95, 2464–2466 (2020).}

_{11. Chang, T.-Y. et al. A High-Precision Deep Learning Algorithm to Localize Idiopathic Ventricular Arrhythmias. J Pers Med 12, 764 (2022).}

_{12. Quartieri, F. et al. Artificial intelligence augments detection accuracy of cardiac insertable cardiac monitors: Results from a pilot prospective observational study. Cardiovasc Digit Health J (2022).}

‍