AI-guided screening technique for Atrial fibrillation improves accuracy: Lancet
A new study conducted by Peter A. Noseworthy and colleagues showed that the yield for atrial fibrillation identification was raised and the efficacy of atrial fibrillation screening may be boosted by using an AI-guided focused screening technique that makes use of current clinical data. The findings of this study were published in The Lancet.
Prior atrial fibrillation screening studies have shown a need for more specialized methods. Researchers conducted a practical investigation to assess the efficacy of a focused screening technique driven by an artificial intelligence (AI) algorithm for detecting atrial fibrillation that had not previously been recognized.
This study prospectively included participants for this non-randomized interventional study who had an electrocardiogram (ECG) performed in a typical setting and who had stroke risk indicators but no known atrial fibrillation. For up to 30 days, participants were required to wear a continuous ambulatory heart rhythm monitor, which provided data almost instantly through a cellular connection. In order to categorize patients into high-risk or low-risk categories, the AI algorithm was applied to the ECGs. Atrial fibrillation that had just been identified was the main result. Propensity-score matched (1:1) trial participants and real-world controls from the eligible but unenrolled population were used in the secondary analysis.
The key findings of this study were:
1. The study was completed by 1003 patients with a mean age of 74 years (SD 8 years) from 40 US.
2. Atrial fibrillation was found in six (1%) of 370 individuals with low risk and 48 (7%) of 633 patients with high risk during the course of a mean 22.3 days of continuous monitoring.
3. Over the course of a median follow-up of 9 months (IQR 7-1-11), AI-guided screening was linked to a higher identification of atrial fibrillation than usual care (high-risk group: 36% [95% CI 23-54] with usual care vs. 106% [8-3-13] with AI-guided screening, p00001; low-risk group: 09% vs. 24%, p=012).
In conclusion, the AI-ECG algorithm opens up the possibility of offering these patients remote monitoring utilizing implanted devices as well as wearable or patch monitors, emphasizing the value of monitoring them for extended periods of time so you can pick up A-fib and in turn also show the influence on clinical outcomes.
Reference:
Noseworthy, P. A., Attia, Z. I., Behnken, E. M., Giblon, R. E., Bews, K. A., Liu, S., Gosse, T. A., Linn, Z. D., Deng, Y., Yin, J., Gersh, B. J., Graff-Radford, J., Friedman, P. A., & Yao, X. (2022). Artificial intelligence-guided screening for atrial fibrillation using electrocardiogram during sinus rhythm: a prospective non-randomised interventional trial. In The Lancet. Elsevier BV. https://doi.org/10.1016/s0140-6736(22)01637-3
