Strategies for Sudden Cardiac Death Prevention

Abstract

Sudden cardiac death (SCD) represents a major challenge in modern medicine. The prevention of SCD orbits on two levels, the general population level and individual level. Much research has been done with the aim to improve risk stratification of SCD, although no radical changes in evidence and in therapeutic strategy have been achieved. Artificial intelligence (AI), and in particular machine learning (ML) models, represent novel technologic tools that promise to improve predictive ability of fatal arrhythmic events. In this review, firstly, we analyzed the electrophysiological basis and the major clues of SCD prevention at population and individual level; secondly, we reviewed the main research where ML models were used for risk stratification in other field of cardiology, suggesting its potentiality in the field of SCD prevention.

Keywords: artificial intelligence; cardiomyopathy; cardiovascular magnetic resonance; machine learning; neural network; risk stratification; sudden cardiac death.

Figure 1

Projecting the growth of publications in PubMed ‘machine learning’. Exponentiated regression of log number of publications on year is used to predict the future trend (adapted from Shameer et al. [3]).

Figure 2

Ionic and molecular basis of cardiac action potential. Left: the ventricular action potential waveform with different phases and representative of inward (blue) or outward (green) currents. Right: molecular components of each ionic current, separated in channel-forming subunits, auxiliary subunits and interacting proteins (adapted from: George A.L. Jr. [5]).

Figure 3

Aspects of cardiovascular magnetic resonance imaging that could be positively impacted by machine learning. These aspects range from patient scheduling to acquisition, image reconstruction, image segmentation, radiomic, classification and prognosis (adapted from Leiner et al. [70]).

Figure 4

The figure retraces the structures of review. Sudden cardiac death (SCD) has been addressed in the two different dimensions of population level and individual level. Current evidence reveals many problems in the accuracy of risk stratification strategies. Solutions proposed in this review suggest the use of ML models to improve patient selection for an implantable cardioverter defibrillator for primary prevention. (a) Adapted from Leiner et al. [64]; (b) lifetime risk for SCD in male population at index age 45 years (left image) and 75 years (right image), adapted from Bogle et al. [6]; (c) annual rate of SCD end point within 5 years stratified according to HCM Risk-SCD), adapted from O’Mahony et al. [50]; (d) adapted from Al’Aref et al. [60].