Predicting the Brugada ECG without ajmaline

Sudden cardiac death accounts for 15% to 20% of all deaths. In cases of sudden unexplained death or cardiac arrest establishing the correct diagnosis is a prerequisite to identifying family members at risk and initiating preventive measures.

In the Brugada syndrome the diagnosis is often established based on the presence of a spontaneous or drug induced (e.g. ajmaline a cardiac sodium channel blocker) Brugada-pattern on the electrocardiogram (ECG). Although complications associated with the procedure are rare, there is a risk of cardiac arrhythmia during the procedure. This project aims at to predicting the occurrence of the drug induced Brugada-pattern using baseline ECGs only.

The project team will use advanced AI methods, such as deep-learning convolutional neural networks, where the features are learned by the network itself, thus potentially allowing the discovery of new ECG features associated with the specific phenotype, genotype or in the case of this project response to a drug challenge.