Herzschrittmachertherapie + Elektrophysiologie

The present article reviews the state of the art of machine learning algorithms for the detection, prediction, and management of atrial fibrillation (AF), as well as of the development and evaluation of artificial intelligence (AI) in cardiology and beyond. Today, AI detects AF with a high accuracy using 12-lead or single-lead electrocardiograms or photoplethysmography. The prediction of paroxysmal or future AF currently operates at a level of precision that is too low for clinical use. Further studies are needed to determine whether patient selection for interventions may be possible with machine learning.

Implantable cardioverter–defibrillators (ICD) have to reliably sense, detect, and treat malignant ventricular tachyarrhythmias. Inappropriate treatment of non life-threatening tachyarrhythmias should be avoided. This article outlines the functionality of ICDs developed and manufactured by BIOTRONIK. Proper sensing is achieved by an automatic sensitivity control which can be individually tailored to solve special under- and oversensing situations. The programming of detection zones for ventricular fibrillation (VF), ventricular tachycardia (VT), and zones to monitor other tachyarrhythmias is outlined. Dedicated single-chamber detection algorithms based on average heart rate, cycle length variability, sudden rate onset, and changes in QRS morphology as used in ICDs by BIOTRONIK are described in detail. Preconditions and confirmation algorithms for therapy deliveries as antitachycardia pacing (ATP) and high energy shocks are explained. Finally, a detailed description of the dual-chamber detection algorithm SMART is given. It comprises additional detection criteria as stability of atrial intervals, 1:1 conduction, atrial–ventricular (AV) multiplicity, AV trend, and AV regularity to differentiate between ventricular and supraventricular tachyarrhythmias.