About the Special Issue

There is growing evidence that current HF classification, based on LV EF, has exhausted its role in the understanding of HF underlying mechanisms and in linking HF types to modern pharmacotherapy. Novel cardiac imaging methods offer meticulous characteristics of cardiac morphology, function, and myocardial tissue. A comprehensive approach utilizing advanced multimodality imaging enables better distinction of cardiac remodeling subtypes and provides additional insights into pathogenesis and clinical manifestations of HF.

Artificial intelligence has a potential to enrich classification of heart failure population into unique entities with different risk profiles and responses to treatment. Cluster analysis methods allow integration of separate imaging parameters into patterns of cardiac remodeling. Machine learning tools enable unsupervised classification of complex variables with nonlinear distribution and confounding factors.

The aim of innovative imaging is to depict pathological processes at molecular level: inflammation, metabolism, apoptosis, necrosis, innervation. Recent trials in conventional heart failure with preserved ejection fraction show an unmet need of alternative phenotyping for designing future clinical studies. Imaging markers can not only guide the choice of HF treatment but are also powerful predictors of patient prognosis. 

Keywords: heart failure; cardiac imaging; phenotyping; cluster analysis; artificial intelligence; machine learning; classification; taxonomy; innovation

Call for Papers