Dear members of the ESC Working Group on e-Cardiology,

It is our pleasure to present another selection of recent publications that show the expanding role of artificial intelligence (AI) and digital technologies in cardiovascular care. Therefore, we recommend these papers to all working group members who want to keep their knowledge in this field up-to-date. 

Our first recommended article, "A wearable-based aging clock associates with disease and behavior by Miller et al". is being extensively discussed by Dr. Michele Orini in the commented article section. Briefly, the authors describe a study with Apple Smartwatches in which a series of PPG signals (each 60 seconds in duration) were used to derive the predicted age, using deep learning models. These signals were collected from 213,593 individual participants. Predicted age strongly correlated with chronological age, behavioural factors, and incident cardiovascular disease. 

The second recommended article is called "Wearable technologies to predict and prevent heart failure hospitalizations: a systematic review" by Noci et al. This paper describes a systematic review following PRISMA guidelines. The authors evaluated the role of non-invasive wearables and looked at the effect on heart failure hospitalisations. They found that heart failure hospitalisations could be predicted 6-32 days in advance. Furthermore, they concluded that therapy based on results from wearables showed a relative reduction of 89% in heart failure hospitalisations after 30 days. This paper, therefore, highlights the potential that wearables have in reducing heart failure hospitalisations. 

Our third article is called "Cardiac amyloidosis detection from a single echocardiographic video clip: a novel artificial intelligence-based screening tool" by Slivnick et al. In this paper, the authors described that they trained a convolutional neural network to be able to distinguish cardiac amyloidosis from other pathologies (such as hypertrophic cardiomyopathy or hypertrophy in case of aortic stenosis) on an apical four-chamber echocardiogram. The authors found a sensitivity of 85% and a specificity of 93% for their model. AI was able to distinguish CA from other causes of hypertrophy and could therefore be a valuable addition in clinical practice to help clinicians recognise CA. 

Together, these papers illustrate the potential of eHealth and artificial intelligence to improve diagnosis and risk prediction and are therefore recommended for reading. 

On behalf of the ESC Working Group on e-Cardiology nucleus, 
Roderick W. Treskes, MD, PhD, MSc 
Secretary, ESC Working Group on e-Cardiology 2024–2026