Long QT syndrome, artificial intelligence, and common sense

This editorial relates to ‘Deep learning analysis of electrocardiogram for risk prediction of drug-induced arrhythmias and diagnosis of long QT syndrome’ by E. Prifti et al., doi:10.1093/eurheartj/ehab588. We are being inundated by so many articles dealing with artificial intelligence (AI) to the point that one begins to wonder about what happened to natural intelligence. Clinical cardiologists are following these rapid developments with mixed feelings. They struggled in the 1990s to survive the assault by the genetic jargon and still shiver recalling how they survived the jungle populated by introns and stop codons, and are now cowed by the new army of ‘machine learning’, ‘convolutional neural networks’, and ‘deep learning algorithms’. These fears, however, are unjustified. Under the cover of Latin, most of these new methodologies are just complex and elegant ways to use a rather old instrument: pattern recognition. This essentially is what we all use to...