Developing digital tools to diagnose and manage atrial fibrillation

Atrial fibrillation, the most common clinical arrhythmia in Sweden, has a substantially increased risk of stroke and death. Patients with atrial fibrillation often have several co-morbidities.

Early detection of atrial fibrillation through screening, risk factor management and rhythm control can reduce the risk for unwanted outcomes in patients with atrial fibrillation.

Screening programs for atrial fibrillation are targeted towards individuals at increased risk of stroke and based on an age cut-off.

From previous screening studies a unique data base of > 400 000 electrocardiograms (ECGs) from 13 000 individuals has been collected.

Using machine learning, we aim to determine if we can increase the precision in determining who will benefit the most from atrial fibrillation screening, both with regards to arrhythmia detection, but also by determining who will be at highest risk for atrial fibrillation-associated mortality and morbidity.

Patients with atrial fibrillation benefit from early heart rhythm management.

In interventional management of atrial fibrillation using ablation increased body mass index and high blood pressure increase the risk of having a repeat procedure.

We propose to co-design a digital tool to intervene on these simple risk factors and study the effects of digital intervention in a randomised controlled trial with the aim to reduce the need for repeat interventions.The results have the potential to impact detection and management of atrial fibrillation.