Abstract Perspective: Long-term PM2.5 Exposure and the Clinical Application of Machine Learning for Predicting Incident Atrial Fibrillation

“Although atrial fibrillation (AF) often leads to complications such as stroke in patients without an awareness of such preexisting diseases, electrocardiogram screening is not sufficient to detect AF in the general population. Some scoring systems for predicting incident AF have been introduced, including the CHADS2, CHA2DS2-VASc, and HATCH scores; however, their prediction accuracies are not sufficient for wide application. Although epidemiological studies have suggested that an elevated level of ambient particulate matter <2.5 μm in aerodynamic diameter (PM2.5) is consistently associated with adverse cardiac events and arrhythmias, including AF, the role of PM2.5 on incident AF remains to be investigated.

We also published about the association between PM2.5 and the incident AF in the general population previously, we then analyzed with machine learning methods in this study and it showed the improved accuracies for predicting incident AF in the general population after applying averaged PM2.5 concentration compared to the existing risk scoring systems.

And finally, we developed the online individual risk calculator for predicting incident AF with applying averaged PM2.5 concentration (https://ml4everyoneisk2.shinyapps.io/RiskCalcAFPM25_ISK). This study was performed with South Korean general population exposed to high levels of air pollution. Further external validation is warranted especially in Western countries affected by low levels of air pollution.”

-Dr. Im-Soo Kim, co-author